http://repub.eur.nl/res/aut/398/ List of Publications en http://repub.eur.nl/static-eur/img/logo.png http://repub.eur.nl http://repub.eur.nl/res/pub/39848/ 2013-04-19T00:00:00Z This paper develops a testing framework for comparing the predictive accuracy of copula-based multivariate density forecasts, focusing on a specific part of the joint distribution. The test is framed in the context of the Kullback-Leibler Information Criterion, but using (out-of-sample) conditional likelihood and censored likelihood in order to focus the evaluation on the region of interest. Monte Carlo simulations document that the resulting test statistics have satisfactory size and power properties in small samples. In an empirical application to daily exchange rate returns we find evidence that the dependence structure varies with the sign and magnitude of returns, such that different parametric copula models achieve superior forecasting performance in different regions of the support. Our analysis highlights the importance of allowing for lower and upper tail dependence for accurate forecasting of common extreme appreciation and depreciation of different currencies. http://repub.eur.nl/res/pub/37989/ 2013-04-01T00:00:00Z This paper compares various ways of extracting macroeconomic information from a data-rich environment for forecasting the yield curve using the Nelson-Siegel model. Five issues in extracting factors from a large panel of macro variables are addressed; namely, selection of a subset of the available information, incorporation of the forecast objective in constructing factors, specification of a multivariate forecast objective, data grouping before constructing factors, and selection of the number of factors in a data-driven way. Our empirical results show that each of these features helps to improve forecast accuracy, especially for the shortest and longest maturities. Factor-augmented methods perform well in relatively volatile periods, including the crisis period in 2008-9, when simpler models do not suffice. The macroeconomic information is exploited best by partial least squares methods, with principal component methods ranking second best. Reductions of mean squared prediction errors of 20-30% are attained, compared to the Nelson-Siegel model without macro factors. http://repub.eur.nl/res/pub/39623/ 2013-04-01T00:00:00Z We introduce a heuristic bias-adjustment for the transaction price-based realized range estimator of daily volatility in the presence of bid-ask bounce and non-trading. The adjustment is an extension of the estimator proposed in Christensen et al. (2009). We relax the assumption that all intraday high (low) transaction prices are at the ask (bid) quote. Using data-based simulations we obtain estimates of the probability that a given intraday range is (upward or downward) biased or not, which we use for a more refined bias-adjustment of the realized range estimator. Both Monte Carlo simulations and an empirical application involving a liquid and a relatively illiquid S&P500 constituent demonstrate that ex post measures and ex ante forecasts based on the heuristically adjusted realized range compare favorably to existing bias-adjusted (two time scales) realized range and (two time scales) realized variance estimators. http://repub.eur.nl/res/pub/38199/ 2012-11-12T00:00:00Z This paper documents that speed is crucially important for high frequency trading strategies based on U.S. macroeconomic news releases. Using order level data of the highly liquid S&P500 ETF traded on NASDAQ from January 6, 2009, to December 12, 2011, we find that a delay of 300 milliseconds (1 second) significantly reduces returns by 3.08% (7.33%) compared to instantaneous execution over all announcements in the sample. This reduction is stronger in case of high impact news and on days with high volatility. In addition, we assess the effect of algorithmic trading on market quality around macroeconomic news. Increases in algorithmic trading activity have a positive (mixed) effect on market quality measures when we use algorithmic trading proxies that capture the top of the orderbook (full orderbook). http://repub.eur.nl/res/pub/37538/ 2012-10-25T00:00:00Z We introduce a heuristic bias-adjustment for the transaction price-based realized range estimator of daily volatility in the presence of bid-ask bounce and non-trading. The adjustment is an extension of the estimator proposed in Christensen et al. (2009). We relax the assumption that all intra-day high (low) transaction prices are at the ask (bid) quote. Using data-based simulations we obtain estimates of the probability that a given intraday range is (upward or downward) biased or not, which we use for a more refined bias-adjustment of the realized range estimator. Both Monte Carlo simulations and an empirical application involving a liquid and a relatively illiquid S&P500 constituent demonstrate that ex post measures and ex ante forecasts based on the heuristically adjusted realized range compare favorably to existing bias-adjusted (two time scales) realized range and (two time scales) realized variance estimators. http://repub.eur.nl/res/pub/37470/ 2012-10-23T00:00:00Z We introduce a Mixed-Frequency Factor Model (MFFM) to estimate vast dimensional covari- ance matrices of asset returns. The MFFM uses high-frequency (intraday) data to estimate factor (co)variances and idiosyncratic risk and low-frequency (daily) data to estimate the factor loadings. We propose the use of highly liquid assets such as exchange traded funds (ETFs) as factors. Prices for these contracts are observed essentially free of microstructure noise at high frequencies, allowing us to obtain precise estimates of the factor covariances. The factor loadings instead are estimated from daily data to avoid biases due to market microstructure effects such as the relative illiquidity of individual stocks and non-synchronicity between the returns on factors and stocks. Our theoretical, simulation and empirical results illustrate that the performance of the MFFM is excellent, both compared to conventional factor models based solely on low-frequency data and to popular realized covariance estimators based on high-frequency data. http://repub.eur.nl/res/pub/34711/ 2012-07-01T00:00:00Z Many economic studies on inflation forecasting have found favorable results when inflation is modeled as a stationary process around a slowly time-varying trend. In contrast, the existing studies on interest rate forecasting either treat yields as being stationary, without any shifting endpoints, or treat yields as a random walk process. In this study we consider the problem of forecasting the term structure of interest rates with the assumption that the yield curve is driven by factors that are stationary around a time-varying trend. We compare alternative ways of modeling the time-varying trend. We find that allowing for shifting endpoints in yield curve factors can provide gains in the out-of-sample predictive accuracy, relative to stationary and random walk benchmarks. The results are both economically and statistically significant. http://repub.eur.nl/res/pub/37672/ 2012-07-01T00:00:00Z We propose a novel approach to active risk management based on the recent Basel II regulations to obtain optimal portfolios with minimum capital requirements. In order to avoid regulatory penalties due to an excessive number of Value-at-Risk (VaR) violations, capital requirements are minimized subject to a given number of violations over the previous trading year. Capital requirements are based on the recent Basel II amendments to account for the 'stressed' VaR, that is, the downside risk of the portfolio under extreme adverse market conditions. An empirical application for two portfolios involving different types of assets and alternative stress scenarios demonstrates that the proposed approach delivers an improved balance between capital requirement levels and the number of VaR exceedances. Furthermore, the risk-adjusted performance of the proposed approach is superior to that of minimum-VaR and minimum-stressed VaR portfolios. http://repub.eur.nl/res/pub/37679/ 2012-05-01T00:00:00Z Institutional investors must deal with irrevocable commitments, cash flow uncertainty, and illiquidity when making new commitments to maintain their portfolio exposure to private equity funds. This study develops a dynamic recommitment strategy to preserve the strategic allocation to private equity. For each period, the level of new commitments is determined by characteristics of the existing private equity portfolio, including received distributions, uncalled capital from old commitments, and the current allocation relative to its target level. [PUBLICATION ABSTRACT] http://repub.eur.nl/res/pub/31778/ 2012-02-01T00:00:00Z This paper investigates the importance of speed for technical trading rule performance for three highly liquid ETFs listed on NASDAQ over the period January 6, 2009 up to September 30, 2009. In addition we examine the characteristics of market activity over the day and within subperiods corresponding to hours, minutes, and seconds. Speed has a clear impact on the return of technical trading rules. For strategies that yield a positive return when they experience no delay, a delay of 200 milliseconds is enough to lower performance significantly. On low volatility days this is already the case for delays larger than 50 milliseconds. In addition, the importance of speed for trading rule performance increases over time. Market activity follows a U-shape over the day with a spike at 10:00AM due to macroeconomic announcements and is characterized by periodic activity within the day, hour, minute, and second. http://repub.eur.nl/res/pub/37834/ 2012-01-01T00:00:00Z Abstract. Managers have the choice to take the firm private themselves in a management buyout or to seek private equity backing. We argue that managers seek private equity backing in case they are more constrained to finance the deal themselves. We confirm the hypothesis using a sample of UK public-to-private transactions over the period 1997–2003. A post going private performance analysis reveals that both management buyouts and private equity backed deals outperform their industry peers. However, private equity backed deals outperform their peers already before the deal takes place whereas management buyouts improve performance afterwards. This suggests a passive role for private equity firms in going private transactions. http://repub.eur.nl/res/pub/26749/ 2012-01-01T00:00:00Z This article addresses heterogeneity in determinants of economic growth in a data-driven way. Instead of defining groups of countries with different growth characteristics a priori, based on, for example, geographical location, we use a finite mixture panel model and endogenous clustering to examine cross-country differences and similarities in the effects of growth determinants. Applying this approach to an annual unbalanced panel of 59 countries in Asia, Latin and Middle America and Africa for the period 1971-2000, we can identify two groups of countries in terms of distinct growth structures. The structural differences between the country groups mainly stem from different effects of investment, openness measures and government share in the economy. Furthermore, the detected segmentation of countries does not match with conventional classifications in the literature. http://repub.eur.nl/res/pub/26863/ 2011-10-01T00:00:00Z This paper examines whether the Conference Board's Leading Economic Index (LEI) can be used for modeling and forecasting a more refined business cycle classification beyond the usual distinction between expansions and contractions. Univariate Markov-switching models for monthly coincident variables and the LEI show that a three regime model is more appropriate than a model with only two regimes. Interestingly, the third regime captures `severe recessions' contrasting the conventional view that the additional third regime represents a 'recovery' phase. This is confirmed by means of Markov-switching vector autoregressive models that allow for phase shifts between the cyclical regimes of LEI and industrial production. Results indicate that a three regime model with a severe recession phase describes the cyclical dynamics in these series better than a two regime model (with only recession and expansion regimes) and a three regime model with a recovery phase. T he timing of the third regime mostly corresponds with periods of substantial credit squeezes and dramatic increases in the default spread as in the recent recession of 2007-2009. These findings provide empirical evidence for the theory of 'financial accelerator'. The severe recession regime of the LEI leads that of IP by 6.5 months whereas for mild recessions this lead time increases to one year. http://repub.eur.nl/res/pub/26086/ 2011-09-02T00:00:00Z This paper develops a novel approach to modeling and forecasting realized volatility (RV) measures based on copula functions. Copula-based time series models can capture relevant characteristics of volatility such as nonlinear dynamics and long-memory type behavior in a flexible yet parsimonious way. In an empirical application to daily volatility for S&P500 index futures, we find that the copula-based RV (C-RV) model outperforms conventional forecasting approaches for one-day ahead volatility forecasts in terms of accuracy and efficiency. Among the copula specifications considered, the Gumbel C-RV model achieves the best forecast performance, which highlights the importance of asymmetry and upper tail dependence for modeling volatility dynamics. Although we find substantial variation in the copula parameter estimates over time, conditional copulas do not improve the accuracy of volatility forecasts. http://repub.eur.nl/res/pub/25708/ 2011-07-13T00:00:00Z This paper examines which macroeconomic and financial variables are most informative for the federal funds target rate decisions made by the Federal Open Market Committee (FOMC) from a forecasting perspective. The analysis is conducted for the FOMC decision during the period January 1990 - June 2008, using dynamic ordered probit models with a Bayesian endogenous variable selection methodology and real-time data for a set of 33 candidate predictor variables. We find that indicators of economic activity and forward-looking term structure variables as well as survey measures have most predictive ability. For the full sample period, in-sample probability forecasts achieve a hitrate of 90 percent. Based on out-of-sample forecasts for the period January 2001 - June 2008, 82 percent of the FOMC decisions are predicted correctly. http://repub.eur.nl/res/pub/26101/ 2011-06-10T00:00:00Z We examine the relation between the quality of corporate governance and the value of excess cash for large publicly listed European firms from common-law and civil-law countries. Besides different law origins, we distinguish different dimensions of corporate governance by using ratings for the quality of Shareholder rights, Takeover defences, Disclosure and Board structure. We find that the value of excess cash is positively related to the Takeover defences score only. This finding is mainly driven by firms from common-law countries. If we focus on changes in the level of excess cash, we do find significant effects for civil-law country firms, where the marginal value of €1 of excess cash in a poorly governed firm is only €0.78 while the value is €1.10 for a good governed firm. We furthermore show that the spending of excess cash by poorly governed firms has a negative impact on their operating performance. http://repub.eur.nl/res/pub/20203/ 2011-06-01T00:00:00Z This paper demonstrates that the Conference Board's Composite Leading Index (CLI) has significant real-time predictive ability for Industrial Production (IP) growth rates at horizons from one to six months ahead over the period 1989-2009. A popular but unrealistic analysis, which combines real-time data for CLI and final vintage data for IP as predictor variables, obscures the actual predictive content of the CLI, in the sense that in that case, the improvements in forecast accuracy relative to a univariate AR model are not significant. The CLI appears to be less useful for forecasting growth rates of the Conference Board's Composite Coincident Index (CCI) in real time, as a univariate AR model performs better. This result is mostly due to its disappointing performance during the first five years of the forecast period. The CLI may not be the best way of exploiting the information contained in the underlying individual leading indicator variables. The use of principal components instead of CLI leads to more accurate real-time forecasts for both IP and CCI growth rates. http://repub.eur.nl/res/pub/22551/ 2011-02-07T00:00:00Z We propose a new approach to deal with structural breaks in time series models. The key contribution is an alternative dynamic stochastic specification for the model parameters which describes potential breaks. After a break new parameter values are generated from a so-called baseline prior distribution. Modeling boils down to the choice of a parametric likelihood specification and a baseline prior with the proper support for the parameters. The approach accounts in a natural way for potential out-of-sample breaks where the number of breaks is stochastic. Posterior inference involves simple computations that are less demanding than existing methods. The approach is illustrated on nonlinear discrete time series models and models with restrictions on the parameter space. http://repub.eur.nl/res/pub/22335/ 2011-01-04T00:00:00Z This paper puts forward kernel ridge regression as an approach for forecasting with many predictors that are related nonlinearly to the target variable. In kernel ridge regression, the observed predictor variables are mapped nonlinearly into a high-dimensional space, where estimation of the predictive regression model is based on a shrinkage estimator to avoid overfitting. We extend the kernel ridge regression methodology to enable its use for economic time-series forecasting, by including lags of the dependent variable or other individual variables as predictors, as is typically desired in macroeconomic and financial applications. Monte Carlo simulations as well as an empirical application to various key measures of real economic activity confirm that kernel ridge regression can produce more accurate forecasts than traditional linear methods for dealing with many predictors based on principal component regression. http://repub.eur.nl/res/pub/22208/ 2011-01-01T00:00:00Z We develop a panel model for regional house prices, for which both the cross-section and the time series dimension is large. The model allows for stochastic trends, cointegration, cross-equation correlations and, most importantly, latent-class clustering of regions. Class membership is fully data-driven and based on the average growth rates of house prices, and the relationship of house prices with economic growth. We apply the model to quarterly data for the Netherlands. The results suggest that there is convincing evidence for the existence of two distinct clusters of regions with pronounced differences in house price dynamics. http://repub.eur.nl/res/pub/25621/ 2011-01-01T00:00:00Z This article documents the existence of large structural breaks in the unconditional correlations among the US dollar exchange rates of the British pound, Norwegian krone, Swedish krona, Swiss franc and euro during the period 1994 to 2003. Using the framework of Dynamic Conditional Correlation (DCC) models, we find that such breaks occurred both at the time the formal decision to proceed with the euro was made in December 1996 and at the time of the actual introduction of the euro in January 1999. Most correlations were substantially lower during the intervening period. We also find breaks in unconditional volatilities at the same points in time, but these are comparatively of a much smaller magnitude. http://repub.eur.nl/res/pub/25629/ 2011-01-01T00:00:00Z A new method of leading index construction is proposed, which explicitly takes into account the purpose of using the index for forecasting a coincident economic indicator. This so-called principal covariate index combines the need for compressing the information in a large number of individual leading indicator variables with the objective of forecasting. In an empirical application to forecast future growth rates of the Conference Board’s Composite Coincident Index and its constituents, the forecasts of the principal covariate index are more accurate than those obtained either from the Composite Leading Index of the Conference Board or from an alternative index-based on principal components. http://repub.eur.nl/res/pub/22327/ 2010-12-01T00:00:00Z This paper develops a Markov-Switching vector autoregressive model that allows for imperfect synchronization of cyclical regimes in multiple variables, due to phase shifts of a single common cycle. The model has three key features: (i) the amount of phase shift can be different across regimes (as well as across variables), (ii) it allows the cycle to consist of any number of regimes J is larger than or equal to 2, and (iii) it allows for regime-dependent volatilities and correlations. In an empirical application to monthly returns on size-based stock portfolios, a three-regime model with asymmetric phase shifts and regime-dependent heteroscedasticity is found to characterize the joint distribution of returns most adequately. While large- and small-cap portfolios switch contemporaneously into boom and crash regimes, the large-cap portfolio leads the small-cap portfolio for switches to a moderate regime by a month. http://repub.eur.nl/res/pub/21861/ 2010-11-01T00:00:00Z This paper documents that factors extracted from a large set of macroeconomic variables bear useful information for predicting monthly US excess stock returns and volatility over the period 1980-2005. Factor-augmented predictive regression models improve upon both benchmark models that only include valuation ratios and interest rate related variables, and possibly individual macro variables, as well as the historical average excess return. The improvements in out-of-sample forecast accuracy are both statistically and economically significant. The factor-augmented predictive regressions have superior market timing ability and volatility timing ability, while a mean-variance investor would be willing to pay an annual performance fee of several hundreds of basis points to switch from the predictions offered by the benchmark models to those of the factor-augmented models. An important reason for the superior performance of the factor-augmented predictive regressions is the stability of their forecast accuracy, whereas the benchmark models suffer from a forecast breakdown during the 1990s. http://repub.eur.nl/res/pub/20622/ 2010-10-01T00:00:00Z We introduce a statistical test for comparing the predictive accuracy of competing copula specifications in multivariate density forecasts, based on the Kullback-Leibler information criterion (KLIC). The test is valid under general conditions on the competing copulas: in particular it allows for parameter estimation uncertainty and for the copulas to be nested or non-nested. Monte Carlo simulations demonstrate that the proposed test has satisfactory size and power properties in finite samples. Applying the test to daily exchange rate returns of several major currencies against the US dollar we find that the Student-t copula is favored over Gaussian, Gumbel and Clayton copulas. http://repub.eur.nl/res/pub/20741/ 2010-09-22T00:00:00Z This paper studies the economic development process, measured by Gross Domestic Product (GDP), for a large panel of countries. We propose a methodology that identifies groups of countries (convergence clubs) that show similar GDP structures, while allowing for changes in club memberships over time. As a second step we analyze the short-run and long-run effects of financial development (measured by financial intermediary development and stock market development) on the GDP process, and the composition of the convergence clubs. We find that the club memberships are quite persistent, but still their compositions change substantially over time. In particular, several EU member countries and East Asian countries are found to belong to a higher GDP club in recent times compared to the beginning of the 1970s. In terms of the effects of financial development indicators on the GDP process, our results partially confirm the theoretical basis for different effects of financial development indicators in the short-run and the long-run. In the long-run, financial development is found to affect the countries’ GDP level positively. The short-run effects of financial development indicators however are found to be less clear, in the sense that we do not find a negative short-run effect of financial intermediary development on GDP levels, while the short-run effect of stock market development is found to be negative. http://repub.eur.nl/res/pub/21017/ 2010-09-01T00:00:00Z We analyse the impact of the Engle and Granger (1987) article by means of its citations over time, and find evidence of a second life starting in the new millennium. Next, we propose a possible explanation of the success of this citation classic. We argue that the conditions for its success were just right at the time of its appearance, because of the growing emphasis on time series properties in econometric modelling, the empirical importance of stochastic trends, the availability of sufficiently long macroeconomic time series, and the availability of personal computers and econometric software for carrying out the new techniques. http://repub.eur.nl/res/pub/19679/ 2010-06-02T00:00:00Z This paper examines the effects of different corporate governance mechanisms on the cost of debt for large European firms and documents a novel interaction effect between shareholder rights and disclosure. Improved disclosure leads to a lower credit spread only if shareholder rights are low. A possible explanation for this finding is the ‘share rights or disclose’ hypothesis. If shareholders have sufficient rights to monitor and influence management decisions, debt providers can rely upon shareholders to mitigate agency costs. Otherwise, bondholders require a premium to compensate for the information risk due to uncertainty about the true value of the firm. http://repub.eur.nl/res/pub/18568/ 2010-04-01T00:00:00Z We study the effects of FOMC announcements of federal funds target rate decisions on individual stock returns, volatilities and correlations at the intraday level. For all three characteristics we find that the stock market responds differently to positive and negative target rate surprises. First, the average response to positive surprises (that is, bad news for stocks) is larger. Second, in case of bad news the mere occurrence of a surprise matters most, whereas for good news its magnitude is more important. These new insights are possible due to the use of high-frequency intraday data. http://repub.eur.nl/res/pub/18254/ 2010-02-23T00:00:00Z Various ways of extracting macroeconomic information from a data-rich environment are compared with the objective of forecasting yield curves using the Nelson-Siegel model. Five issues in factor extraction are addressed, namely, selection of a subset of the available information, incorporation of the forecast objective in constructing factors, specification of a multivariate forecast objective, data grouping before constructing factors, and selection of the number of factors in a data-driven way. Our empirical results show that each of these features helps to improve forecast accuracy, especially for the shortest and longest maturities. The data-driven methods perform well in relatively volatile periods, when simpler models do not suffice. http://repub.eur.nl/res/pub/18571/ 2010-02-01T00:00:00Z Using an Euler discretization to simulate a mean-reverting CEV process gives rise to the problem that while the process itself is guaranteed to be nonnegative, the discretization is not. Although an exact and efficient simulation algorithm exists for this process, at present this is not the case for the CEV-SV stochastic volatility model, with the Heston model as a special case, where the variance is modelled as a mean-reverting CEV process. Consequently, when using an Euler discretization, one must carefully think about how to fix negative variances. Our contribution is threefold. Firstly, we unify all Euler fixes into a single general framework. Secondly, we introduce the new full truncation scheme, tailored to minimize the positive bias found when pricing European options. Thirdly and finally, we numerically compare all Euler fixes to recent quasi-second order schemes of Kahl and Jckel, and Ninomiya and Victoir, as well as to the exact scheme of Broadie and Kaya. The choice of fix is found to be extremely important. The full truncation scheme outperforms all considered biased schemes in terms of bias and root-mean-squared error http://repub.eur.nl/res/pub/17169/ 2009-11-01T00:00:00Z This paper shows that stock market contagion occurs as a domino effect, where confined local crashes evolve into more widespread crashes. Using a novel framework based on ordered logit regressions we model the occurrence of local, regional and global crashes as a function of their past occurrences and financial variables. We find significant evidence that global crashes do not occur abruptly but are preceded by local and regional crashes. Besides this form of contagion, interdependence shows up by the effect of interest rates, bond returns and stock market volatility on crash probabilities. When it comes to forecasting global crashes, our model outperforms a binomial model for global crashes only. http://repub.eur.nl/res/pub/17096/ 2009-10-20T00:00:00Z The dependence between asset returns varies. Its strength can become stronger or weaker. Also, its structure can change, for example, when asymmetries related to bull and bear markets become more or less pronounced. To analyze these different types of variations, we develop a model that separately accommodates these changes. It combines a mixture of structurally different copulas with time variation. Our model shows both types of changes in the dependence between several equity market returns. Ignoring them leads to biases in risk measures. An underestimation of Value-at-Risk by maximum 15% occurs exactly when most harmful, during crisis periods. http://repub.eur.nl/res/pub/17018/ 2009-10-19T00:00:00Z Macroeconomic forecasting is not an easy task, in particular if future growth rates are forecasted in real time. This paper compares various methods to predict the growth rate of US Industrial Production (IP) and of the Composite Coincident Index (CCI) of the Conference Board, over the coming month, quarter, and half year. It turns out that future IP growth rates can be forecasted in real time from ten leading indicators, by means of the Composite Leading Index (CLI) or, even somewhat better, by principal components regression. This amends earlier negative findings for IP by Diebold and Rudebusch. For CCI, on the other hand, simple autoregressive models do not provide significantly less accurate forecasts than single-equation and bivariate vector autoregressive models with the CLI. This amends some of the more positive results for CCI recently reported by the Conference Board. Not surprisingly, all forecast methods improve considerably if `ex post' data are used, after possible data updates and revisions. http://repub.eur.nl/res/pub/17214/ 2009-10-08T00:00:00Z We introduce the realized co-range, a novel estimator of the daily covariance between asset returns based on intraday high-low price ranges. In an ideal world, the co-range is five times more efficient than the realized covariance, which uses cross-products of intraday returns, when sampling at the same frequency. In Monte Carlo simulations, we find that for plausible levels of bid-ask bounce, infrequent trading and nonsynchronous trading, the realized co-range still improves upon the realized covariance. In a volatility timing strategy for S&P500, bond and gold futures, we find that the co-range estimates are less noisy, which results in lower transaction costs and higher Sharpe ratios. http://repub.eur.nl/res/pub/16035/ 2009-06-01T00:00:00Z We measure the economic value of information derived from macroeconomic variables and from technical trading rules for emerging markets currency investments. Our analysis is based on a sample of 21 emerging markets with a floating exchange rate regime over the period 1997-2007 and explicitly accounts for trading restrictions on foreign capital movements by using non-deliverable forward data. We document that both types of information can be exploited to implement profitable trading strategies. In line with evidence from surveys of foreign exchange professionals concerning the use of fundamental and technical analysis, we find that combining the two types of information improves the risk-adjusted performance of the investment strategies. http://repub.eur.nl/res/pub/15779/ 2009-05-07T00:00:00Z We analyse the impact of the Engle and Granger (1987) article by its citations over time, and find evidence of a second life starting in the new millennium. Next, we propose a possible explanation of the success of this citation classic. We argue that the conditions for its success were just right at the time of its appearance, because of the growing emphasis on time-series properties in econometric modelling, the empirical importance of stochastic trends, the availability of sufficiently long macro-economic time series, and the availability of personal computers and econometric software to carry out the new techniques. http://repub.eur.nl/res/pub/18152/ 2009-04-01T00:00:00Z We evaluate the forecasting performance of time series models for realized volatility, which accommodate long memory, level shifts, leverage effects, day-of-the-week and holiday effects, as well as macroeconomic news announcements. Applying the models to daily realized volatility for the S&P 500 futures index, we find that explicitly accounting for these stylized facts of volatility improves out-of-sample forecast accuracy for horizons up to 20 days ahead. Capturing the long memory feature of realized volatility by means of a flexible high-order AR-approximation instead of a parsimonious but stringent fractionally integrated specification also leads to improvements in forecast accuracy, especially for longer horizon forecasts. http://repub.eur.nl/res/pub/18164/ 2009-04-01T00:00:00Z This Special Issue brings together a selection of the papers presented at the third conference in the Economic and Social Research Council (ESRC) Seminar series “Nonlinear Economics and Finance Research Community”, as well as a number of other related contributions. This Conference took place at Keele University (UK) on the 1st of February 2008, and was hosted by Christopher Martin (Brunel University), Costas Milas (Keele University) and Theodore Panagiotidis (University of Macedonia), with funding from the ESRC under grant RES-451-25-4260. The aim of the seminar series is to bring together researchers working on nonlinear topics in economics and finance http://repub.eur.nl/res/pub/16383/ 2009-01-01T00:00:00Z This paper provides a comprehensive analysis of the functional form of the euro-area Phillips curve over the past three decades. In particular, compared with previous literature, we analyze the stability of the relationship in detail, especially as regards the possibility of a time-varying mean of inflation. Moreover, we conduct a sensitivity analysis across different measures of economic slack. Our main findings are two. First, there is strong evidence of time variation in the mean and slope of the Phillips curve occurring in the early to mid-1980s, but not in inflation persistence once the mean shift is allowed for. As a result of the structural change, the Phillips curve became flatter around a lower mean of inflation. Second, we find no significant evidence of nonlinearity - in particular, in relation to the output gap. http://repub.eur.nl/res/pub/13835/ 2008-11-05T00:00:00Z This paper shows that stock market contagion operates through a domino effect, where small crashes evolve into more severe crashes. Using a novel unifying framework we model the occurrence of local, regional and global crashes in terms of past occurrences of these different crashes and financial variables. We find convincing evidence that global crashes do not occur abruptly but are preceded by local and regional crashes. Additionally, interest rates, bond returns and volatility affect the probabilities of different crash types, indicating interdependence. We show that in forecasting global crashes our model outperforms a binomial model for global crashes only. http://repub.eur.nl/res/pub/14052/ 2008-11-01T00:00:00Z We introduce a statistical test for comparing the predictive accuracy of competing copula specifications in multivariate density forecasts, based on the Kullback-Leibler Information Criterion (KLIC). The test is valid under general conditions: in particular it allows for parameter estimation uncertainty and for the copulas to be nested or non-nested. Monte Carlo simulations demonstrate that the proposed test has satisfactory size and power properties in finite samples. Applying the test to daily exchange rate returns of several major currencies against the US dollar we find that the Student's t copula is favored over Gaussian, Gumbel and Clayton copulas. This suggests that these exchange rate returns are characterized by symmetric tail dependence. http://repub.eur.nl/res/pub/14044/ 2008-08-29T00:00:00Z This paper addresses heterogeneity in determinants of economic growth in a data-driven way. Instead of defining groups of countries with different growth characteristics a priori, based on, for example, geographical location, we use a finite mixture panel model and endogenous clustering to examine cross-country differences and similarities in the effects of growth determinants. Applying this approach to an annual unbalanced panel of 59 countries in Asia, Latin and Middle America and Africa for the period 1971-2000, we can identify two groups of countries in terms of distinct growth structures. The structural differences between the country groups mainly stem from different effects of investment, openness measures and government share in the economy. Furthermore, the detected segmentation of countries does not match with conventional classifications in the literature. http://repub.eur.nl/res/pub/12465/ 2008-05-20T00:00:00Z We examine the relation between the quality of corporate governance and the value of excess cash for large European firms (FTSEurofirst 300 Index). We use Deminor ratings for Shareholder rights, Takeover defences, Disclosure and Board as proxies for the quality of corporate governance. We find that the value of excess cash is positively related to the Takeover defences score only. It seems that governance mechanisms—except the market for corporate control—are not strong enough to prevent managers from wasting excess cash. For non-UK firms we find that the value of €1 of excess cash in a poorly governed firm is valued at only €0.89 while the value is €1.45 for a good governed firm. We show that poorly governed firms dissipate excess cash relatively quickly with a negative impact on their operating performance as a result. http://repub.eur.nl/res/pub/13975/ 2008-05-19T00:00:00Z We propose new scoring rules based on partial likelihood for assessing the relative out-of-sample predictive accuracy of competing density forecasts over a specific region of interest, such as the left tail in financial risk management. By construction, existing scoring rules based on weighted likelihood or censored normal likelihood favor density forecasts with more probability mass in the given region, rendering predictive accuracy tests biased towards such densities. Our novel partial likelihood-based scoring rules do not suffer from this problem, as illustrated by means of Monte Carlo simulations and an empirical application to daily S&P 500 index returns. http://repub.eur.nl/res/pub/11556/ 2008-03-03T00:00:00Z This paper presents empirical evidence that security analysts do not efficiently use publicly available macroeconomic information in their earnings forecasts for emerging market stocks. Analysts completely ignore forecasts on political stability, while these provide valuable information for firm-level earnings growth. Analysts do incorporate output growth forecasts, but these actually bear no relevant information for firm-level earnings growth. Inflation forecasts are taken into account correctly. In addition, the information environment appears to be crucially important in emerging markets, as we find evidence that analysts handle macroeconomic information in a better way for more transparent firms. http://repub.eur.nl/res/pub/18645/ 2008-02-01T00:00:00Z This article proposes an improved method for the construction of principal components in macroeconomic forecasting. The underlying idea is to maximize the amount of variance of the original predictor variables that is retained by the components in order to reduce the variance involved in estimating the forecast model. This is achieved by matching the data window used for constructing the components with the estimation window. Extensive Monte Carlo simulations, using dynamic factor models, clarify the relationship between the achieved reduction in forecast variance and various design parameters, such as the observation length, the number of predictors, and the length of the forecast horizon. The method is also used in an empirical application to forecast eight key US macroeconomic time series over various horizons, where the components are constructed from a large set of predictors. The results show that the proposed modification leads, on average, to more accurate forecasts than previously used principal component regression methods. http://repub.eur.nl/res/pub/10904/ 2008-01-15T00:00:00Z We introduce the realized co-range, utilizing intraday high-low price ranges to estimate asset return covariances. Using simulations we find that for plausible levels of bid-ask bounce and infrequent and non-synchronous trading the realized co-range improves upon the realized covariance, which uses cross-products of intraday returns. One advantage of the co-range is that in an ideal world it is five times more efficient than the realized covariance when sampling at the same frequency. The second advantage is that the upward bias due to bid-ask bounce and the downward bias due to infrequent and non-synchronous trading partially offset each other. In a volatility timing strategy for S\\&P500, bond and gold futures we find that the co-range estimates are less noisy as exemplified by lower transaction costs and also higher Sharpe ratios when using more weight on recent data for predicting covariances. http://repub.eur.nl/res/pub/18655/ 2008-01-01T00:00:00Z This article investigates the merits of high-frequency intraday data when forming mean-variance efficient stock portfolios with daily rebalancing from the individual constituents of the S&P 100 index. We focus on the issue of determining the optimal sampling frequency as judged by the performance of these portfolios. The optimal sampling frequency ranges between 30 and 65 minutes, considerably lower than the popular five-minute frequency, which typically is motivated by the aim of striking a balance between the variance and bias in covariance matrix estimates due to market microstructure effects such as non-synchronous trading and bid-ask bounce. Bias-correction procedures, based on combining low-frequency and high-frequency covariance matrix estimates and on the addition of leads and lags do not substantially affect the optimal sampling frequency or the portfolio performance. Our findings are also robust to the presence of transaction costs and to the portfolio rebalancing frequency. http://repub.eur.nl/res/pub/10892/ 2007-12-24T00:00:00Z This paper develops a reinvestment strategy for private equity which aims to keep its portfolio weight equal to a desired strategic allocation, while taking into account the illiquid nature of private equity. Historical simulations (1980-2005) show that our dynamic strategy is capable of maintaining a stable investment level that is close to the target. This does not only hold for unrestricted portfolios, but also for investments limited to buyout or venture capital, a specific region, or management experience. This finding is of great importance for investors, because private equity funds have a finite lifetime and uncertain cash flows. http://repub.eur.nl/res/pub/10891/ 2007-12-21T00:00:00Z We measure the economic value of information derived from macroeconomic variables and from technical trading rules for emerging markets currency investments. Our analysis is based on a sample of 21 emerging markets with a floating exchange rate regime over the period 1997-2007 and explicitly accounts for trading restrictions on foreign capital movements by using non-deliverable forward data. We document that both types of information can be exploited to implement profitable trading strategies. In line with evidence from surveys of foreign exchange professionals concerning the use of fundamental and technical analysis, we find that combining the two types of information improves the risk-adjusted performance of the investment strategies. http://repub.eur.nl/res/pub/11723/ 2007-12-01T00:00:00Z We develop a parsimonious panel model for quarterly regional house prices, for which both the cross-section and the time series dimension is large. The model allows for stochastic trends, cointegration, cross-equation correlations and, most importantly, latent-class clustering of regions. Class membership is fully data-driven and based on (i) average growth rates of house prices, (ii) the propagation of shocks to house prices across regions, also known as the ripple effect, and (iii) the relationship of house prices with economic growth and other variables. Applying the model to quarterly data for the Netherlands, we find convincing evidence for the existence of two distinct clusters of regions, with pronounced differences in house price dynamics. http://repub.eur.nl/res/pub/10673/ 2007-11-15T00:00:00Z Toen u de titel van deze rede voor de eerste keer las, vroeg u zich misschien af of het vanmiddag wel over econometrie zou gaan, aangezien dat woord in het geheel niet in de titel voorkomt. Echter, met een echtgenote en een schoonvader die beiden in het Hilversumse Mediapark werken is het mij inmiddels wel duidelijk geworden dat een titel allesbepalend is voor de grootte van het publiek dat je trekt. Als de titel van je programma geen eye-catcher is hoef je bij voorbaat al niet te rekenen op bijzonder hoge kijkcijfers. Vandaar "Goed nieuws is geen nieuws". Gezien het feit dat u hier vanmiddag bent heeft die titel blijkbaar de gewenste uitwerking gehad, en heeft die u nieuwsgierig genoeg gemaakt om de reis naar Rotterdam te ondernemen. Rede, in verkorte vorm uitgesproken op donderdag 15 november 2007 bij de aanvaarding van het ambt van hoogleraar aan de Faculteit der Economische Wetenschappen van de Erasmus Universiteit Rotterdam met de leeropdracht Econometrie (Finance). http://repub.eur.nl/res/pub/10857/ 2007-11-15T00:00:00Z News plays a crucial role in determining prices in financial markets. In an efficient market, current prices fully and correctly reflect all available information, such that only truly new information leads to price adjustment. This lecture shows that using high-frequency data makes it possible to accurately measure the reaction of stock prices on the New York stock exchange to new information related to the Federal funds target rate. An unexpected change in the target rate of 25 basis points leads to a return of slightly more than one percent within five minutes after the news announcement. Furthermore, the effects of positive and negative news on stock prices are fundamentally different. In case of positive news the stock market reaction depends upon the magnitude of the unexpected decrease of the interest rate; in case of negative news, stock prices only respond to the fact that an unexpected rate increase occurs. http://repub.eur.nl/res/pub/10610/ 2007-10-25T00:00:00Z We study the impact of FOMC announcements of Federal funds target rate decisions on individual stock prices at the intraday level. We find that the returns, volatilities and correlations of the S&P100 index constituents only respond to the surprise component in the announcement, as measured by the change in the Federal funds futures rate. For example, an unexpected 25 basis points increase of the target rate leads on average to a 113 basis points negative market return within five minutes after the announcement. It also increases market volatility during the 60-minute window around the announcement with 147 basis points. Positive surprises, meaning bad news for stocks, provoke a stronger reaction than negative surprises. Market participants also respond differently to good and bad news. In case of bad news for stocks the fact that there is a surprise matters most, whereas in case of good news the magnitude of the surprise is more important. Across sectors, Financials and IT show the strongest response to target rate surprises. http://repub.eur.nl/res/pub/10467/ 2007-08-27T00:00:00Z The accuracy of real-time forecasts of macroeconomic variables that are subject to revisions may crucially depend on the choice of data used to compare the forecasts against. We put forward a flexible time-varying parameter regression framework to obtain early estimates of the final value of macroeconomic variables based upon the initial data release that may be used as actuals in current forecast evaluation. We allow for structural changes in the regression parameters to accommodate benchmark revisions and definitional changes, which fundamentally change the statistical properties of the variable of interest, including the relationship between the final value and the initial release. The usefulness of our approach is demonstrated through an empirical application comparing the accuracy of forecasts of US GDP growth rates from the Survey of Professional Forecasters and the Greenbook. http://repub.eur.nl/res/pub/11249/ 2007-08-27T00:00:00Z Dataset in Excel consisting of three parts: dataset 1, dataset 2, and forecasts. http://repub.eur.nl/res/pub/11120/ 2007-05-15T00:00:00Z It generally is difficult, if not impossible, to fully understand and interpret nonlinear time series models by considering the estimated values of the model parameters only. To shed light on the characteristics and implications of a nonlinear model it can then be useful to consider the effects of shocks on the future patterns of the time series variable. Most interest in such impulse response analysis has concentrated on measuring the persistence of shocks, or the magnitude of their (ultimate) effect. A framework is developed and implemented that is useful for measuring the rate at which this final effect is attained, or the rate of absorption of shocks. It is shown that the absorption rate can be used to examine whether the propagation of different types of shocks, such as positive and negative shocks or large and small shocks follows different patterns. The nonlinear floor-and-ceiling model for US output growth is used to illustrate the various concepts. The presence of substantial asymmetries in both persistence and absorption of shocks is documented, with interesting differences arising across magnitudes of shocks and across regimes in the model. Furthermore, it appears that asymmetry became much less pronounced due to a large decline in output volatility in the 1980s. http://repub.eur.nl/res/pub/10070/ 2007-05-10T00:00:00Z Over the last decade, the going private market has experienced a considerable boom in size and also has become more interesting for private equity investors that are looking to partner with incumbent management. This offers managers the choice to take the firm private themselves in a traditional management buyout or to seek private equity backing. We propose that managers decide for a management buyout without any involvement of private equity in case they are less financially constrained: when their firms are undervalued, have high cash levels, are smaller and less financially visible, and the managers own a large toehold. In contrast, managers invite participation of private equity investors when they cannot complete the deal themselves: in firms that are larger, have less cash and managers own a smaller fraction of the firm. Our analysis on a sample of UK public-to-private transactions completed over the period 1997-2003 provides results that are in line with these predictions. http://repub.eur.nl/res/pub/11121/ 2007-05-01T00:00:00Z Realized variance, being the summation of squared intra-day returns, has quickly gained popularity as a measure of daily volatility. Following Parkinson [1980. The extreme value method for estimating the variance of the rate of return. Journal of Business 53, 61–65] we replace each squared intra-day return by the high–low range for that period to create a novel and more efficient estimator called the realized range. In addition, we suggest a bias-correction procedure to account for the effects of microstructure frictions based upon scaling the realized range with the average level of the daily range. Simulation experiments demonstrate that for plausible levels of non-trading and bid–ask bounce the realized range has a lower mean-squared error than the realized variance, including variants thereof that are robust to microstructure noise. Empirical analysis of the S&P500 index-futures and the S&P100 constituents confirms the potential of the realized range. http://repub.eur.nl/res/pub/11122/ 2007-04-01T00:00:00Z Forecasting with many predictors is of interest, for instance, in macroeconomics and finance. The forecast accuracy of two methods for dealing with many predictors is compared, that is, principal component regression (PCR) and principal covariate regression (PCovR). Simulation experiments with data generated by factor models and regression models indicate that, in general, PCR performs better for the first type of data and PCovR performs better for the second type of data. An empirical application to four key US macroeconomic variables shows that PCovR achieves improved forecast accuracy in some situations. http://repub.eur.nl/res/pub/9230/ 2007-03-20T00:00:00Z We develop a formal statistical approach to investigate the possibility that leading indicator variables have different lead times at business cycle peaks and troughs. For this purpose, we propose a novel Markov switching vector autoregressive model, where economic growth and leading indicators share a common Markov process determining the state, but such that their cycles are non-synchronous with the non-synchronicity varying across the different regimes. An empirical application to monthly US industrial production (IP) and The Conference Board's Composite Index of Leading Indicators (CLI) for the period 1959-2004 shows that on average the CLI leads IP by more than seven months at peaks, but only by three and a half months at troughs. In terms of timeliness, the CLI is therefore most useful for signalling oncoming recessions. Furthermore, we find that allowing for asymmetric lead times leads to improved real-time dating of business cycle peaks and troughs and more accurate forecasts of turning points and IP growth. http://repub.eur.nl/res/pub/9148/ 2007-03-03T00:00:00Z We forecast the term structure of U.S. Treasury zero-coupon bond yields by analyzing a range of models that have been used in the literature. We assess the relevance of parameter uncertainty by examining the added value of using Bayesian inference compared to frequentist estimation techniques, and model uncertainty by combining forecasts from individual models. Following current literature we also investigate the benefits of incorporating macroeconomic information in yield curve models. Our results show that adding macroeconomic factors is very beneficial for improving the out-of-sample forecasting performance of individual models. Despite this, the predictive accuracy of models varies over time considerably, irrespective of using the Bayesian or frequentist approach. We show that mitigating model uncertainty by combining forecasts leads to substantial gains in forecasting performance, especially when applying Bayesian model averaging. http://repub.eur.nl/res/pub/11123/ 2007-03-01T00:00:00Z This paper demonstrates that the class of conditionally linear and Gaussian state-space models offers a general and convenient framework for simultaneously handling nonlinearity, structural change and outliers in time series. Many popular nonlinear time series models, including threshold, smooth transition and Markov-switching models, can be written in state-space form. It is then straightforward to add components that capture parameter instability and intervention effects. We advocate a Bayesian approach to estimation and inference, using an efficient implementation of Markov Chain Monte Carlo sampling schemes for such linear dynamic mixture models. The general modelling framework and the Bayesian methodology are illustrated by means of several examples. An application to quarterly industrial production growth rates for the G7 countries demonstrates the empirical usefulness of the approach. http://repub.eur.nl/res/pub/8003/ 2006-08-31T00:00:00Z This paper is concerned with time series forecasting in the presence of a large number of predictors. The results are of interest, for instance, in macroeconomic and financial forecasting where often many potential predictor variables are available. Most of the current forecast methods with many predictors consist of two steps, where the large set of predictors is first summarized by means of a limited number of factors -for instance, principal components- and, in a second step, these factors and their lags are used for forecasting. A possible disadvantage of these methods is that the construction of the components in the first step is not directly related to their use in forecasting in the second step. This motivates an alternative method, principal covariate regression (PCovR), where the two steps are combined in a single criterion. This method has been analyzed before within the framework of multivariate regression models. Moti- vated by the needs of macroeconomic time series forecasting, this paper discusses two adjustments of standard PCovR that are necessary to allow for lagged factors and for preferential predictors. The resulting nonlinear estimation problem is solved by means of a method based on iterative majorization. The paper discusses some numerical aspects and analyzes the method by means of simulations. Further, the empirical per- formance of PCovR is compared with that of the two-step principal component method by applying both methods to forecast four US macroeconomic time series from a set of 132 predictors, using the data set of Stock and Watson (2005). http://repub.eur.nl/res/pub/7904/ 2006-08-24T00:00:00Z This paper develops a return forecasting methodology that allows for instabil ity in the relationship between stock returns and predictor variables, for model uncertainty, and for parameter estimation uncertainty. The predictive regres sion speci¯cation that is put forward allows for occasional structural breaks of random magnitude in the regression parameters, and for uncertainty about the inclusion of forecasting variables, and about the parameter values by em ploying Bayesian Model Averaging. The implications of these three sources of uncertainty, and their relative importance, are investigated from an active investment management perspective. It is found that the economic value of incorporating all three sources of uncertainty is considerable. A typical in vestor would be willing to pay up to several hundreds of basis points annually to switch from a passive buy-and-hold strategy to an active strategy based on a return forecasting model that allows for model and parameter uncertainty as well as structural breaks in the regression parameters. http://repub.eur.nl/res/pub/11124/ 2006-06-20T00:00:00Z A response surface analysis for the distributions of popular tests for seasonal unit roots in quarterly observed time series variables is presented. Five test statistics are considered, along with the most commonly used specifications of the deterministic component in the test regression; allowance is also made for the lag order in the test regression to be determined endogenously, using commonly applied selection methods. Response surface coefficients are reported, permitting simple computation of accurate critical values for 1%-, 5%- and 10%-level tests and probability values for any sample size and lag order. Accurate approximations of the asymptotic distributions are obtained in the process of constructing the response surfaces. Dependence of the critical values and the probability density functions on the sample size and lag order is investigated. http://repub.eur.nl/res/pub/7738/ 2006-05-17T00:00:00Z When using an Euler discretisation to simulate a mean-reverting square root process, one runs into the problem that while the process itself is guaranteed to be nonnegative, the discretisation is not. Although an exact and efficient simulation algorithm exists for this process, at present this is not the case for the Heston stochastic volatility model, where the variance is modelled as a square root process. Consequently, when using an Euler discretisation, one must carefully think about how to fix negative variances. Our contribution is threefold. Firstly, we unify all Euler fixes into a single general framework. Secondly, we introduce the new full truncation scheme, tailored to minimise the upward bias found when pricing European options. Thirdly and finally, we numerically compare all Euler fixes to a recent quasi-second order scheme of Kahl and Jäckel and the exact scheme of Broadie and Kaya. The choice of fix is found to be extremely important. The full truncation scheme by far outperforms all biased schemes in terms of bias, root-mean-squared error, and hence should be the preferred discretisation method for simulation of the Heston model and extensions thereof. http://repub.eur.nl/res/pub/7581/ 2006-02-28T00:00:00Z This article proposes a modified method for the construction of diffusion indexes in macroeconomic forecasting using principal component regres- sion. The method aims to maximize the amount of variance of the origi- nal predictor variables retained by the diffusion indexes, by matching the data windows used for constructing the principal components and for es- timating the diffusion index models. The method is applied to construct forecasts of eight monthly US macroeconomic time series, using the data set of Stock and Watson (2002a). The results show that the proposed method leads, on average, to simpler models with smaller forecast errors than previously used methods. http://repub.eur.nl/res/pub/7582/ 2006-02-28T00:00:00Z Realized variance, being the summation of squared intra-day returns, has quickly gained popularity as a measure of daily volatility. Following Parkinson (1980) we replace each squared intra-day return by the high-low range for that period to create a novel and more efficient estimator called the realized range. In addition we suggest a bias-correction procedure to account for the effects of microstructure frictions based upon scaling the realized range with the average level of the daily range. Simulation experiments demonstrate that for plausible levels of non-trading and bid-ask bounce the realized range has a lower mean squared error than the realized variance, including variants thereof that are robust to microstructure noise. Empirical analysis of the S&P500 index-futures and the S&P100 constituents confirm the potential of the realized range. http://repub.eur.nl/res/pub/11125/ 2006-01-15T00:00:00Z The so-called Harrod–Balassa–Samuelson model implies that relative prices of non-traded goods may be nonstationary and, hence, that PPP should preferably be tested on real exchange rates based on prices of traded goods only. A simple test for PPP among traded goods is proposed that can be applied to real exchange rates based on prices of all (that is, both traded and non-traded) goods. The study shows through simulations that the test is reliable for a sample size commonly considered in practice. Upon applying the test to bilateral real exchange rates based on the general CPI among a group of industrialized countries during the post Bretton Woods period, we find little evidence in favour of PPP among traded goods. This does not change when we use real exchange rates based on various components of the CPI. http://repub.eur.nl/res/pub/11162/ 2006-01-01T00:00:00Z http://repub.eur.nl/res/pub/11170/ 2006-01-01T00:00:00Z http://repub.eur.nl/res/pub/11131/ 2005-11-01T00:00:00Z Causality-in-variance tests suffer from severe size distortions in the presence of structural breaks in volatility, when such breaks are not taken into account. Pre-testing the series for structural changes in volatility largely remedies the problem. http://repub.eur.nl/res/pub/11134/ 2005-10-01T00:00:00Z Macroeconomic time series such as total unemployment or total industrial production concern data which are aggregated across regions, sectors, or age categories. In this paper we examine whether forecasts for these aggregates can be improved by considering panel models for the disaggregate series. As many macroeconomic variables have nonlinear properties, we specifically focus on panels of nonlinear time series. We discuss the representation of such models, parameter estimation and a method for generating forecasts. We illustrate the usefulness of our approach for simulated data and for the US coincident index, making use of state-specific component series. http://repub.eur.nl/res/pub/11137/ 2005-10-01T00:00:00Z In this paper, we examine the forecast accuracy of linear autoregressive, smooth transition autoregressive (STAR), and neural network (NN) time series models for 47 monthly macroeconomic variables of the G7 economies. Unlike previous studies that typically consider multiple but fixed model specifications, we use a single but dynamic specification for each model class. The point forecast results indicate that the STAR model generally outperforms linear autoregressive models. It also improves upon several fixed STAR models, demonstrating that careful specification of nonlinear time series models is of crucial importance. The results for neural network models are mixed in the sense that at long forecast horizons, an NN model obtained using Bayesian regularization produces more accurate forecasts than a corresponding model specified using the specific-to-general approach. Reasons for this outcome are discussed. http://repub.eur.nl/res/pub/6959/ 2005-09-21T00:00:00Z This paper investigates the merits of high-frequency intraday data when forming minimum variance portfolios and minimum tracking error portfolios with daily rebalancing from the individual constituents of the S&P 100 index. We focus on the issue of determining the optimal sampling frequency, which strikes a balance between variance and bias in covariance matrix estimates due to market microstructure effects such as non-synchronous trading and bid-ask bounce. The optimal sampling frequency typically ranges between 30- and 65-minutes, considerably lower than the popular five-minute frequency. We also examine how bias-correction procedures, based on the addition of leads and lags and on scaling, and a variance-reduction technique, based on subsampling, affect the performance. http://repub.eur.nl/res/pub/11140/ 2005-09-01T00:00:00Z We examine competing explanations, based on risk and behavioral models, for the profitability of stock selection strategies in emerging markets. We document that both emerging market risk and global risk factors cannot account for the significant excess returns of selection strategies based on value, momentum and earnings revisions indicators. The findings for value and momentum strategies are consistent with the evidence from developed markets supporting behavioral explanations. In addition, for value stocks, the most important behavioral bias appears to be related to underestimation of long-term growth prospects, as indicated by above average earnings revisions for longer post-formation horizons and by quite rapidly improving earnings growth expectations. Furthermore, we find that overreaction effects play a limited role for the earnings revisions strategy, as there is no clear return reversal until five years after portfolio formation, setting this strategy apart from momentum strategies. http://repub.eur.nl/res/pub/13409/ 2005-09-01T00:00:00Z We introduce a multi-level smooth transition model for a panel of time series, which can be used to examine the presence of common nonlinear business cycle features across many variables. The model is positioned in between a fully pooled model, which imposes such common features, and a fully heterogeneous model, which allows for unrestricted nonlinearity. We introduce a second-stage model linking the parameters that determine the timing of the switches between business cycle regimes to observable explanatory variables, thereby allowing for lead-lag relationships across panel members. We discuss representation, estimation by concentrated simulated maximum likelihood and inference. We illustrate our model using quarterly industrial production in 19 US manufacturing sectors, and document that there are subtle differences across sectors in leads and lags for switches between business cycle recessions and expansions. http://repub.eur.nl/res/pub/6918/ 2005-08-02T00:00:00Z Forecasting with many predictors is of interest, for instance, in macroeconomics and finance. This paper compares two methods for dealing with many predictors, that is, principal component regression (PCR) and principal covariate regression (PCovR). The forecast performance of these methods is compared by simulating data from factor models and from regression models. The simulations show that, in general, PCR performs better for the first type of data and PCovR performs better for the second type of data. The simulations also clarify the effect of the choice of the PCovR weight on the orecast quality. http://repub.eur.nl/res/pub/11142/ 2005-08-01T00:00:00Z We address the question whether sub-Saharan African countries have lower average growth rates in real GDP per capita than countries in Asia, Latin and Middle America and the Middle East. In contrast to previous studies, countries are no a priori assigned to clusters based on geographical location. Instead, we propose a latent-class panel time series model, which allows a data-based classification of countries into clusters such that within a cluster countries have the same average growth rate. Our empirical results suggest that three clusters are sufficient to describe the different growth paths. Twenty-six African countries belong to the low growth cluster, but 8 African countries show growth rates comparable with many countries in Asia, Latin and Middle America and the Middle East. http://repub.eur.nl/res/pub/11377/ 2005-07-01T00:00:00Z http://repub.eur.nl/res/pub/6849/ 2005-07-01T00:00:00Z In this paper we develop a new semi-parametric model for conditional correlations, which combines parametric univariate GARCH-type specifications for the individual conditional volatilities with nonparametric kernel regression for the conditional correlations. This approach not only avoids the proliferation of parameters as the number of assets becomes large, which typically happens in conventional multivariate conditional volatility models, but also the rigid structure imposed by more parsimonious models, such as the dynamic conditional correlation model. An empirical application to the 30 Dow Jones stocks demonstrates that the model is able to capture interesting asymmetries in correlations and that it is competitive with standard parametric models in terms of constructing minimum variance portfolios and minimum tracking error portfolios. http://repub.eur.nl/res/pub/6583/ 2005-04-01T00:00:00Z This paper documents the existence of large structural breaks in the unconditional correlations among the British pound, Norwegian krone, Swedish krona, Swiss franc, and euro exchange rates (against the US dollar) during the period 1994-2003. Using the framework of dynamic conditional correlation (DCC) models, we find that such breaks occurred both at the time the formal decision to proceed with the euro was made in December 1996 and at the time of the actual introduction of the euro in January 1999. In particular, we document that most correlations were substantially lower during the intermittent period. We also find breaks in unconditional volatilities at the same points in time, but these are of a much smaller magnitude comparatively. http://repub.eur.nl/res/pub/1922/ 2005-03-29T00:00:00Z We examine competing explanations, based on risk and behavioral models, for the profitability of stock selection strategies in emerging markets. We document that both emerging market risk and global risk factors cannot account for the significant excess returns of selection strategies based on value, momentum and earnings revisions indicators. The findings for value and momentum strategies are consistent with the evidence from developed markets supporting behavioral explanations. In addition, for value stocks, the most important behavioral bias appears to be related to underestimation of long-term growth prospects, as indicated by overly pessimistic analysts' earnings forecasts and above average earnings revisions for longer postformation horizons and by quite rapidly improving earnings growth expectations. Furthermore, we find that overreaction effects play a limited role for the earnings revisions strategy, as there is no clear return reversal up until five years after portfolio formation, setting this strategy apart from momentum strategies. http://repub.eur.nl/res/pub/1910/ 2005-03-09T00:00:00Z This paper demonstrates that the class of conditionally linear and Gaussian state-space models offers a general and convenient framework for simultaneously handling nonlinearity, structural change and outliers in time series. Many popular nonlinear time series models, including threshold, smooth transition and Markov-Switching models, can be written in state-space form. It is then straightforward to add components that capture parameter instability and intervention effects. We advocate a Bayesian approach to estimation and inference, using an efficient implementation of Markov Chain Monte Carlo sampling schemes for such linear dynamic mixture models. The general modelling framework and the Bayesian methodology are illustrated by means of several examples. An application to quarterly industrial production growth rates for the G7 countries demonstrates the empirical usefulness of the approach. http://repub.eur.nl/res/pub/11143/ 2005-01-01T00:00:00Z Seasonality often accounts for the major part of short-run movements in quarterly or monthly macro economic time series. In addition, business cycle nonlinearity is a prominent feature of many such series. A forecaster can nowadays consider a wide variety of time series models that describe seasonal variation and nonlinear regime-switching behavior. In this paper we examine the forecasting performance of various models for seasonality and nonlinearity for quarterly industrial production series of 18 OECD countries. We find that the accuracy of point forecasts varies widely across series, across forecast horizons and across seasons. However, in general, linear models with fairly simple descriptions of seasonality outperform nonlinear at short forecast horizons, whereas nonlinear models with more elaborate seasonal components dominate at longer horizons. Finally, none of the models is found to render efficient forecasts and hence, forecast combination is worthwhile. http://repub.eur.nl/res/pub/1797/ 2004-11-05T00:00:00Z Macroeconomic time series such as total unemployment or total industrial production concern data which are aggregated across regions, sectors, or age categories. In this paper we examine if forecasts for these aggregates can be improved by considering panel models for the disaggregate series. As many macroeconomic variables have nonlinear properties, we specifically focus on panels of nonlinear time series. We discuss the representation of such models, parameter estimation and a method to generate forecasts. We illustrate the usefulness of our approach for simulated data and for the US coincident index, making use of state-specific component series. http://repub.eur.nl/res/pub/1801/ 2004-11-05T00:00:00Z We examine the size properties of tests for causality in variance in the presence of structural breaks in volatility. Extensive Monte Carlo simulations demonstrate that these tests suffer from severe size distortions when such breaks are not taken into account. Pre-testing the series for structural changes in volatility is shown to largely remedy the problem. http://repub.eur.nl/res/pub/1627/ 2004-09-22T00:00:00Z We consider tests for sudden changes in the unconditional volatility of conditionally heteroskedastic time series based on cumulative sums of squares. When applied to the original series these tests suffer from severe size distortions, where the correct null hypothesis of no volatility change is rejected much too frequently. Applying the tests to standardized residuals from an estimated GARCH model results in good size and reasonable power properties when testing for a single break in the variance. The tests also appear to be robust to different types of misspecification. An iterative algorithm is designed to test sequentially for the presence of multiple changes in volatility. An application to emerging markets stock returns clearly illustrates the properties of the different test statistics. http://repub.eur.nl/res/pub/11144/ 2004-08-01T00:00:00Z We test for a change in the volatility of 214 U.S. macroeconomic time series over the period 1959–1999. We find that approximately 80% of these series have experienced a break in unconditional volatility during this period. Even though more than half of the series experienced a break in conditional mean, most of the reduction in volatility appears to be due to changes in conditional volatility. Our results are robust to controlling for business cycle nonlinearity in both mean and variance. Volatility changes are more appropriately characterized as instantaneous breaks than as gradual changes. Nominal variables such as inflation and interest rates experienced multiple volatility breaks and witnessed temporary increases in volatility during the 1970s. On this evidence, we conclude that the increased stability of economic fluctuations is widespread. http://repub.eur.nl/res/pub/6630/ 2004-06-05T00:00:00Z The sum of squared intraday returns provides an unbiased and almost error-free measure of ex-post volatility. In this paper we develop a nonlinear Autoregressive Fractionally Integrated Moving Average (ARFIMA) model for realized volatility, which accommodates level shifts, day-of-the-week effects, leverage effects and volatility level effects. Applying the model to realized volatilities of the S&P 500 stock index and three exchange rates produces forecasts that clearly improve upon the ones obtained from a linear ARFIMA model and from conventional time-series models based on daily returns, treating volatility as a latent variable. http://repub.eur.nl/res/pub/6645/ 2004-05-27T00:00:00Z This paper considers financial, operational, solvency, and performance ratios, in order to detect when there were balance sheets’ variations related to the 1994 Mexican currency crisis. Quarterly results for 88 non-financial Mexican companies that survived the crisis are used, and tests for structural change are performed. Findings show that generally firms’ balance sheets deteriorated between the fourth quarters of 1993 and 1995, which points the possibility of corporate roots of the macroeconomic crisis. Although in most cases firms’ balance sheets improved after the crisis, the recovery was partial and gradual, and overall this episode was prejudicial even for surviving companies. http://repub.eur.nl/res/pub/2179/ 2004-02-01T00:00:00Z The (Generalized) AutoRegressive Conditional Heteroscedasticity [(G)ARCH] model is tested for daily data on 22 exchange rates and 13 stock market indices using the standard Lagrange Multiplier [LM] test for GARCH and a LM test that is resistant to patches of additive outliers. The data span two samples of five years ranging from 1986 to 1995. Using asymptotic arguments and Monte Carlo simulations, in which the empirical method is evaluated, it is shown that patches of outliers can have significant effects on test outcomes. The main empirical result is that spurious GARCH is found in about 40% of the cases, while in many other cases evidence of GARCH is found even though such sequences of extraordinary observations seem to be present. http://repub.eur.nl/res/pub/11163/ 2003-12-01T00:00:00Z Nonlinear time series models have become fashionable tools to describe and forecast a variety of economic time series. A closer look at reported empirical studies, however, reveals that these models apparently fit well in-sample, but rarely show a substantial improvement in out-of-sample forecasts, at least over linear models. One of the many possible reasons for this finding is the use of inappropriate model selection criteria and forecast evaluation criteria. In this paper we therefore propose a novel criterion, which we believe does more justice to the very nature of nonlinear models. Simulations show that this criterion outperforms those criteria currently in use, in the sense that the true nonlinear model is more often found to perform better in out-of-sample forecasting than a benchmark linear model. An empirical illustration for US GDP emphasizes its relevance. http://repub.eur.nl/res/pub/11145/ 2003-08-01T00:00:00Z We compare linear autoregressive (AR) models and self-exciting threshold autoregressive (SETAR) models in terms of their point forecast performance, and their ability to characterize the uncertainty surrounding those forecasts, i.e. interval or density forecasts. A two-regime SETAR process is used as the data-generating process in an extensive set of Monte Carlo simulations, and we consider the discriminatory power of recently developed methods of forecast evaluation for different degrees of non-linearity. We find that the interval and density evaluation methods are unlikely to show the linear model to be deficient on samples of the size typical for macroeconomic data. http://repub.eur.nl/res/pub/1716/ 2003-05-14T00:00:00Z We compare the forecasting performance of linear autoregressive models, autoregressive models with structural breaks, self-exciting threshold autoregressive models, and Markov switching autoregressive models in terms of point, interval, and density forecasts for h-month growth rates of industrial production of the G7 countries, for the period January 1960-December 2000. The results of point forecast evaluation tests support the established notion in the forecasting literature on the favorable performance of the linear AR model. By contrast, the Markov switching models render more accurate interval and density forecasts than the other models, including the linear AR model. This encouraging finding supports the idea that non-linear models may outperform linear competitors in terms of describing the uncertainty around future realizations of a time series. http://repub.eur.nl/res/pub/1703/ 2003-03-26T00:00:00Z Nonlinear time series models have become fashionable tools to describe and forecast a variety of economic time series. A closer look at reported empirical studies, however, reveals that these models apparently fit well in-sample, but rarely show a substantial improvement in out-of-sample forecasts, at least over linear models. One of the many possible reasons for this finding is that inappropriate model selection criteria and forecast evaluation criteria are used. In this paper we therefore propose a novel criterion, which we believe does more justice to the very nature of nonlinear models. Simulations show that our criterion outperforms currently used criteria, in the sense that the true nonlinear model is more often found to perform better in out-of-sample forecasting than a benchmark linear model. An empirical illustration for US GDP emphasizes its relevance. http://repub.eur.nl/res/pub/11147/ 2003-02-01T00:00:00Z We examine the profitability of a broad range of stock selection strategies in 32 emerging markets over the period 1985–1999. Value, momentum and earnings revisions strategies are most successful and generate significant excess returns, in contrast to strategies based on size, liquidity and mean reversion. The performance of the strategies can be enhanced by selecting stocks on multiple characteristics and by incorporating country selection, although the latter bears the cost of increased risk. We do not find a pronounced effect of financial market liberalization on the performance of the strategies. There is no evidence that global risk factors can account for the excess returns of selection strategies. Finally, we document that the strategies can be implemented successfully in practice by a large institutional investor, facing a lack of liquidity and substantial transaction costs. http://repub.eur.nl/res/pub/1054/ 2003-01-01T00:00:00Z We introduce a multi-level smooth transition model for a panel of time series variables, which can be used to examine the presence of common non-linear features across many such variables. The model is positioned in between a fully pooled model, which imposes such common features, and a fully heterogeneous model, which might render estimation problems for some of the panel members. To keep the model tractable, we introduce a second-stage model, which links the parameters in the transition functions with observable explanatory variables. We discuss representation, estimation by concentrated simulated maximum likelihood and inference. We illustrate our model for data on industrial production of 18 US manufacturing sectors, and document that there are subtle differences across sectors in leads and lags for business cycle recessions and expansions. http://repub.eur.nl/res/pub/11146/ 2003-01-01T00:00:00Z Changes in the seasonal patterns of macroeconomic time series may be due to the effects of business cycle fluctuations or to technological and institutional change or both. We examine the relative importance of these two sources of change in seasonality for quarterly industrial production series of the G7 countries using time-varying smooth transition autoregressive models. We find compelling evidence that the effects of gradual institutional and technological change are much more important than the effects attributable to the business cycle. http://repub.eur.nl/res/pub/11148/ 2003-01-01T00:00:00Z Nonlinear regime-switching behavior and structural change are often perceived as competing alternatives to linearity. In this article we study the so-called time-varying smooth transition autoregressive (TV-STAR) model, which can be used both for describing simultaneous nonlinearity and structural change and for distinguishing between these features. Two modeling strategies for empirical specification of TV-STAR models are developed. Monte Carlo simulations show that neither of the two strategies dominates the other. A specific-to-general-to-specific procedure is best suited for obtaining a first impression of the importance of nonlinearity and/or structural change for a particular time series. A specific-to-general procedure is most useful in careful specification of a model with nonlinear and/or time-varying properties. An empirical application to a large dataset of U.S. macroeconomic time series illustrates the relative merits of both modeling strategies. http://repub.eur.nl/res/pub/11175/ 2003-01-01T00:00:00Z http://repub.eur.nl/res/pub/11176/ 2003-01-01T00:00:00Z http://repub.eur.nl/res/pub/1695/ 2003-01-01T00:00:00Z In this paper we address the question whether countries on the African continent have lower average growth rates in real GDP per capita than countries in Asia and Latin America. In contrast to previous studies, we do not aggregate the data, nor do we a priori assign countries to clusters. Instead, we put forward a so-called latent class panel time series model, which allows a data-based classification of countries to clusters with growth levels that differ across the clusters. Our empirical results suggest that twenty-six African countries can be assigned to the low growth cluster, but that eleven African countries show growth levels which are comparable with many countries in Asia and Latin America. We also present results for sub-periods, which demonstrate that the relative performance of African countries has improved considerably over time. http://repub.eur.nl/res/pub/11150/ 2002-10-01T00:00:00Z Two important empirical features of US unemployment are that shocks to the series seem rather persistent and that it seems to rise faster during recessions than that it falls during expansions. To jointly capture these features of long memory and nonlinearity, we put forward a new time series model and evaluate its empirical performance. We find that the model describes the data rather well and that it outperforms related competitive models on various measures of fit. http://repub.eur.nl/res/pub/551/ 2002-09-19T00:00:00Z Volatility breaks are tested and documented for 19 important monthly macroeconomic time series across the G7 countries. Across all conditional mean specifications considered, including both linear and nonlinear models with and without a structural break, volatility breaks are found to be widespread. This continues to hold when business cycle nonlinearities are allowed in the variance. Multiple volatility breaks are also examined, and these are found to be especially prevalent for short-term interest rates. Volatility breaks in industrial production and consumer prices are largely synchronous across the G7. The facts established are discussed in the context of some explanations put forward in the literature to explain volatility breaks previously found for US series. http://repub.eur.nl/res/pub/11149/ 2002-01-01T00:00:00Z In this paper we investigate whether or not the recently developed class of tests of the unit root null against the alternative of a stochastic unit root forms a useful statistical tool in distinguishing between time series processes whose degree of persistence is no more than that of a unit root [I(1)] process and those which display a greater degree of persistence than I(1) series, the stochastic unit root process being an example of the latter. For a wide range of processes which have been put forward as serious competitors to the I(1) process, both of a greater and lesser degree of persistence, we find, via numerical simulation methods, that broadly speaking the stochastic unit root tests do indeed appear to provide an efficacious diagnostic tool in this regard. http://repub.eur.nl/res/pub/11151/ 2002-01-01T00:00:00Z This paper surveys recent developments related to the smooth transition autoregressive (STAR) time series model and several of its variants. We put emphasis on new methods for testing for STAR nonlinearity, model evaluation, and forecasting. Several useful extensions of the basic STAR model, which concern multiple regimes, time-varying non-linear properties, and models for vector time series, are also reviewed. http://repub.eur.nl/res/pub/585/ 2002-01-01T00:00:00Z The so-called Balassa-Samuelson model implies that relative prices of non-traded goods may be nonstationary and, hence, that PPP should preferably be tested on real exchange rates based on prices of traded goods only. We propose a simple test for PPP among traded goods which can be applied to real exchange rates based on prices of all (that is, both traded and non-traded) goods. We show through simulations that the test is reliable for a sample size commonly considered in practice. Upon applying the test to bilateral real exchange rates based on the general CPI among a group of industrialized countries during the recent float, we find little evidence in favor of PPP among traded goods. This does not change when we use real exchange rates based on various components of the CPI. http://repub.eur.nl/res/pub/1694/ 2001-09-28T00:00:00Z This paper provides new evidence on the rationality of industrial production (IP) and the producer price index (PPI). However, rather than examining preliminary and fully revised data, as is usually the practice, we examine the entire revision history for each data series. Thus, we are able to assess whether earlier releases of data are in any sense "less" rational than later releases, for example, and when early releases of data become rational. Our findings suggest that seasonally unadjusted IP and PPI become rational after approximately 3-4 months, while seasonally adjusted versions of these series remain irrational for at least 12 months after initial release. Additionally, we find that there is a clear increase in the volatility of early data releases during recessions, suggesting that early data are less reliable in tougher economic times. One feature of the approach that we take is that we are able to include revision histories in the information sets used to examine the rationality of a particular release of data. This in turn allows us to assess whether the revision process itself is predictable from its own past, hence possibly leading to rules for the construction of "better" preliminary releases of data. For most of the variables examined, we find evidence of this form of predictability. Another feature of the approach taken in the paper is that we are able to provide evidence suggesting that nonlinearities in economic behavior manifest themselves in the form of nonlinearities in the rationality of early releases of economic data. This is done by separately analyzing expansionary and recessionary economic phases and by allowing for structural breaks. These types of nonlinearities are shown to be prevalent, and in some cases incorrect inferences concerning unbiasedness and efficiency arise when they are not taken account of. For example, seasonally unadjusted IP data become unbiased much more quickly after 1980 than before 1980. Additionally, seasonally adjusted IP data take less time to become efficient during expansions than during recessions. http://repub.eur.nl/res/pub/1678/ 2001-04-26T00:00:00Z Seasonality often accounts for the major part of quarterly or monthly movements in detrended macro-economic time series. In addition, business cycle nonlinearity is a prominent feature of many such series too. A forecaster can nowadays consider a wide variety of time series models which describe seasonal variation and regime-switching behaviour. In this paper we examine the forecasting performance of various models for seasonality and nonlinearity using quarterly industrial production series for 17 OECD countries. We find that forecasting performance varies widely across series, across forecast horizons and across seasons. However, in general, linear models with fairly simple descriptions of seasonality outperform at short forecast horizons, whereas nonlinear models with more elaborate seasonal components dominate at longer horizons. http://repub.eur.nl/res/pub/1674/ 2001-03-30T00:00:00Z We test for a change in the volatility of 215 US macroeconomic time series over the period 1960-1996. We find that about 90\\% of these series have experienced a break in volatility during this period. This result is robust to controlling for instability in the mean and business cycle nonlinearities. Real variables have seen a reduction in volatility since the early 1980s, which is accompanied by lower but steadier output growth. Furthermore, nominal variables have seen temporary increases in their volatility around the early 1980s. This suggests the existence of a trade-off between short-term volatility and the long-term pattern of growth. http://repub.eur.nl/res/pub/1676/ 2001-03-30T00:00:00Z Changes in the seasonal patterns of macroeconomic time series may be due to the effects of business cycle fluctuations or to technological and institutional change or both. We examine the relative importance of these two sources of change in seasonality for industrial production series of the G7 countries. We find compelling evidence that the effects of gradual institutional and technological change are much more important than the effects attributable to the business cycle. http://repub.eur.nl/res/pub/11153/ 2001-03-13T00:00:00Z In addition to clear-cut seasonality in mean and variance, weekly Dutch temperature data appear to have a strong asymmetry in the impact of unexpectedly high or low temperatures on conditional volatility. Furthermore, this asymmetry also shows fairly pronounced seasonal variation. To describe these features, we propose a univariate seasonal time series model with asymmetric conditionally heteroskedastic errors. We fit this (and other, nested) model(s) to 25 years of weekly data. We evaluate its forecasting performance for 5 years of hold-out data and find that the imposed asymmetry leads to better out-of-sample forecasts of temperature volatility. http://repub.eur.nl/res/pub/6879/ 2001-01-26T00:00:00Z Recent empirical evidence suggests that value and momentum strategies generate significant excess returns in emerging markets. We confirm these results and extend them in several directions. First, we examine a broader range of stock selection strategies, including strategies based on analysts' earnings revisions. We also consider multivariate strategies, whereby stocks are selected on multiple characteristics, and find that this enhances the overall performance. Excess returns also increase if country selection is incorporated into the strategies, but the risk of the strategies increases proportionally. Second, we test whether the strategies can be implemented successfully in practice by a large institutional investor, facing a lack of liquidity, restrictions on foreign ownership and substantial transaction costs. We find that even under such more realistic circumstances the strategies earn significant excess returns. Third, we examine several popular explanations for the excess returns. We find no evidence of higher market risk or lower liquidity of the strategies. Instead, based on the developments of earnings and earnings revisions after portfolio formation, we find that the results are consistent with behavioral explanations. http://repub.eur.nl/res/pub/1660/ 2000-10-05T00:00:00Z Two important empirical features of monthly US unemployment are that shocks to the series seem rather persistent and that unemployment seems to rise faster in recessions than that it falls during expansions. To jointly capture these features of long memory and nonlinearity, respectively, we put forward a new time series model and evaluate its empirical performance. We find that the model describes the data rather well and that it outperforms related competitive models on various measures of fit. http://repub.eur.nl/res/pub/2175/ 2000-08-01T00:00:00Z This paper provides an empirical description of the relationship between the trading system operated by a stock exchange and the trading behaviour of heterogeneous investors who use the exchange. The recent introduction of SETS in the London Stock Exchange provides an excellent opportunity to study the impact of an electronic trading system upon traders who use the exchange. Using the cost-of-carry model of futures prices we estimate (non-linearly) the transaction costs and trade speeds faced by arbitragers who take advantage of mispricing of FTSE100 futures contracts relative to the spot prices of the stocks that make up the FTSE100 stock index. We divide the sample period into pre-SETS and post-SETS sample periods and conduct a comparative study of arbitrager behaviour under different trading systems. The results indicate that there has been a significant reduction in the level of transaction costs faced by arbitragers and in the degree of transaction cost heterogeneity. Finally, generalised impulse response functions show that both spot and futures prices adjust more quickly in the post-SETS period. These results suggest that both spot and futures markets have become more efficient under SETS. http://repub.eur.nl/res/pub/1656/ 2000-06-09T00:00:00Z This paper surveys recent developments related to the smooth transition autoregressive [STAR] time series model and several of its variants. We put emphasis on new methods for testing for STAR nonlinearity, model evaluation, and forecasting. Several useful extensions of the basic STAR model, which concern multiple regimes, time-varying nonlinear properties, and models for vector time series, are also reviewed. http://repub.eur.nl/res/pub/1639/ 2000-02-04T00:00:00Z In this paper we put forward a new time series model, which describes nonlinearity and seasonality simultaneously. We discuss its representation, estimation of the parameters and inference. This seasonal STAR (SEASTAR) model is examined for its practical usefulness by applying it to 18 quarterly industrial production series. The data are tested for smooth-transition nonlinearity and for time-varying seasonality. We find that the model fits the data well for 14 of the 18 series. We also consider out-of-sample forecasting where we compare forecasts from the SEASTAR models with forecasts from nested models. It turns out that the SEASTAR model sometimes outperforms the other models, particularly for large horizons. Finally, we compare the SEASTAR models with STAR models for the 14 corresponding seasonally adjusted series, and we find that the estimated business cycle chronologies can be markedly different. http://repub.eur.nl/res/pub/1637/ 2000-01-20T00:00:00Z A key feature of many nonlinear time series models is that they allow for the possibility that the model structure experiences changes, depending on for example the state of the economy or of the financial market. A common property of these models is that it generally is not possible to fully understand the structure of the model by considering the estimated values of the model parameters only. Put differently, it often is difficult to interpret a specific nonlinear model. To shed light on the characteristics of a nonlinear model it can then be useful to consider the effect of shocks on the future patterns of a time series variable. Most interest in such impulse response analysis has concentrated on measuring the persistence of shocks, or the magnitude of the (ultimate) effect of shocks. Interestingly, far less attention has been given to measuring the speed at which this final effect is attained, that is, how fast shocks are 'absorbed' by a time series. In this paper we develop and implement a framework that can be used to assess the absorption rate of shocks in nonlinear models. The current-depth-of-recession model of Beaudry and Koop (1993), the floor-and-ceiling model of Pesaran and Potter (1997) and a multivariate STAR model are used to illustrate the various concepts. http://repub.eur.nl/res/pub/2125/ 2000-01-01T00:00:00Z http://repub.eur.nl/res/pub/1616/ 1999-11-12T00:00:00Z Using a standard 4-variable linear vector error correction model (VECM), we first show that the null hypothesis of linearity can be strongly rejected against the alternative of smooth transition autoregressive nonlinearity. An important result from this stage of the analysis is that the quarterly growth rate of money is identified as the transition variable, the variable which governs the smooth switching between regimes. This implies there is a nonlinear causal relationship between money and output. A smooth transition VECM (STVECM) is then used to examine whether money nonlinearly Granger causes output in the sense that lagged values of money enter the model's output equation as regressors. We evaluate this type of nonlinear Granger causality with both in-sample and out-of-sample analysis. For the in-sample analysis we compare alternative models using predictive accuracy tests. These results vary strongly across use of the AIC and SIC. Our use of an out-of-sample forecasting exercise to study money-income Granger causality, both linear and nonlinear, we believe is new to the literature. The forecasting results do not suggest that money is nonlinearly Granger causal for output. In fact, they show that by allowing money to nonlinearly Granger cause output, the forecasting performance of the STVECM is significantly worsened. http://repub.eur.nl/res/pub/1856/ 1999-09-16T00:00:00Z The dynamic properties of many economic time series variables can be characterised as state-dependent or regime-switching. A popular model to describe this type of non-linear behaviour is the smooth transition model, which accommodates two regimes facilitating a gradual transition from one regime to the other. The first part of this thesis considers three extensions of the basic smooth transition model. Models are developed which allow for more than two regimes, for time-varying properties in conjunction with regime-switching behaviour, and for modeling several time series jointly. Particular emphasis is placed on the inter-related issues of specification and inference in such models. The second part of the thesis concerns the influence of atypical observations on testing procedures for smooth transition non-linearity and on the estimation of smooth transition models. Traditional methods that are used for these purposes are found to be very sensitive to such outliers. Therefore, outlier robust testing procedures and estimation methods are developed http://repub.eur.nl/res/pub/1597/ 1999-07-05T00:00:00Z In this paper the issue of detecting and handling outliers in the GARCH(1,1) model is addressed. Simulation evidence shows that neglecting even a single outlier has a dramatic on parameter estimates. To detect and correct for outliers, we propose an adaptation of the iterative in Chen and Liu (1993, JASA). We generate the critical values for the relevant test statistic, and we evaluate our method in an extensive simulation study. An application to several weekly stock return series shows that correcting for a few outliers yields substantial improvements in out-of-sample forecasts. http://repub.eur.nl/res/pub/1592/ 1999-05-26T00:00:00Z This paper considers the properties of the tests of the null hypothesis of a random walk against the alternative of a stochastic unit root when the data are generated by a process related to, but not exactly equal to, the processes under the null and alternative. The extensive Monte Carlo experiments demonstrate that the test statistics are particularly sensitive to non-stationary RCAR processes with mean root less than unity, random walk processes with structural change in the variance, processes with changing persistence and trend-stationary processes with a break in the trend. http://repub.eur.nl/res/pub/1558/ 1999-03-29T00:00:00Z Standard unit root tests are misspecified in case the variable of interest is stationary but displays asymmetric adjustment towards its long-run equilibrium and, consequently, may suffer from a lack of power against such alternatives. This observation recently has aroused interest in developing test statistics which can be used to test the null hypothesis of a unit root against the alternative of stationarity with asymmetric adjustment. In this paper we reconsider the test statistics put forward by Enders and Granger (1998). We point out an important deficiency of their tests and develop an alternative one which is based on more solid statistical grounds. Monte Carlo experiments demonstrate that our new test outperforms standard unit roots and the tests of Enders and Granger (1998) in terms of power against the alternative of interest. An empirical illustration involving the forward premium is provided to demonstrate the practical usefulness of our test statistic. http://repub.eur.nl/res/pub/11154/ 1999-01-01T00:00:00Z In this paper we investigate the properties of the Lagrange Multiplier [LM] test for autoregressive conditional heteroscedasticity (ARCH) and generalized ARCH (GARCH) in the presence of additive outliers (AOs). We show analytically that both the asymptotic size and power are adversely affected if AOs are neglected: the test rejects the null hypothesis of homoscedasticity too often when it is in fact true, while the test has difficulty detecting genuine GARCH effects. Several Monte Carlo experiments show that these phenomena occur in small samples as well. We design and implement a robust test, which has better size and power properties than the conventional test in the presence of AOs. We apply the tests to a number of US macroeconomic time series, which illustrates the dangers involved when nonrobust tests for ARCH are routinely applied as diagnostic tests for misspecification. http://repub.eur.nl/res/pub/11157/ 1999-01-01T00:00:00Z Regime-switching models, like the smooth transition autoregressive (STAR) model, are typically applied to time series of moderate length. Hence, the nonlinear features that these models intend to describe may be reflected in only a few observations. Conversely, neglected outliers in a linear time series of moderate length may incorrectly suggest STAR (or other) types of nonlinearity. Outlier robust tests are proposed for STAR-type nonlinearity. These tests are designed such that they have a better level and power behavior than standard nonrobust tests in situations with outliers. Local and global robustness properties of the new tests are formally derived. Extensive Monte Carlo simulations show the practical usefulness of the robust tests. An application to several quarterly industrial production indexes illustrates that apparent nonlinearity in time series sometimes seems due to only a few outliers. http://repub.eur.nl/res/pub/7729/ 1999-01-01T00:00:00Z This paper provides an empirical description of the relationship between the trading system operated by a stock exchange and the transaction costs faced by heterogeneous investors who use the exchange. The recent introduction of SETS in the London Stock Exchange provides an excellent opportunity to study the impact of an electronic trading system upon transaction costs and the time taken to carry out a trade. Using the cost-of-carry model of futures prices we estimate (non-linearly) the transaction costs and trade speeds faced by arbitragers who take advantage of mispricing of FTSE100 futures contracts relative to the spot prices of the stocks that make up the FTSE100 stock index. We divide the sample period into pre-SETS and post-SETS sample periods and conduct a comparative study of arbitrager behaviour under different trading systems. The results indicate that there has been a significant reduction in the level of transaction costs faced by arbitragers and in the degree of transaction cost heterogeneity since the introduction of SETS. Finally, generalised impulse response functions show that both spot and futures prices adjust more quickly in the post-SETS period. http://repub.eur.nl/res/pub/1555/ 1998-12-31T00:00:00Z One of the stylized facts in financial and international economics is that of increasing predictability of variables such as exchange rates and stock returns at longer horizons. This fact is based upon applications of long horizon regressions, from which the typical findings are that the point estimates of the regression parameter, the associated t-statistic, and the regression R^2 all tend to increase as the horizon increases. Such long horizon regression analyses implicitly assume the existence of cointegration between the variables involved. In this paper, we investigate the consequences of dropping this assumption. In particular, we look upon the long horizon regression as a conditional error-correction model and interpret the test for long horizon predictability as a single equation test for cointegration. We derive the asymptotic distributions of the estimator of the regression parameter and its t-statistic for arbitrary horizons, under the null hypothesis of no cointegration. It is shown that these distributions provide an alternative explanation for at least part of the typical findings. Furthermore, the distributions are used to derive a Phillips-Perron type correction to the ordinary least-squares t-statistic in order to endow it with a stable size for given, arbitrary, horizon. A local asymptotic power analysis reveals that the power of long horizon regression tests does not increase with the horizon. Exchange rate data are used to demonstrate the empirical relevance of our theoretical results. http://repub.eur.nl/res/pub/1533/ 1998-09-21T00:00:00Z In addition to clear-cut seasonality in mean and variance, weekly Dutch temperature data appear to have a strong asymmetry in the impact of unexpectedly high or low temperatures on conditional volatility. Furthermore, this asymmetry also shows fairly pronounced seasonal variation. To describe these features, we propose a univariate seasonal time series model with asymmetric conditionally heteroskedastic errors. We fit this (and other, nested) model(s) to 25 years of weekly data. We evaluate its forecasting performance for 5 years of hold-out data and find that the imposed asymmetry leads to better out-of-sample forecasts of temperature volatility. http://repub.eur.nl/res/pub/1542/ 1998-08-10T00:00:00Z Outliers and nonlinearity may easily be mistaken. This paper uses Monte Carlo methods to examine and compare the behavior of two competing specification procedures for Smooth Transition AutoRegressive [STAR] models under various different circumstances (linear and nonlinear data generating processes, with and without outlier contamination). The extensive simulation evidence demonstrates that the use of outlier-robust variants of the linearity tests which are involved leads to procedures with more desirable properties. An application to several real exchange rate series illustrates the potential usefulness of the robust specification procedures, especially in case one is not certain whether or not aberrant observations are present. http://repub.eur.nl/res/pub/1553/ 1998-06-16T00:00:00Z In this paper we examine the forecasting performance of five nonlinear GARCH(1,1) models. Four of these have recently been proposed in literature, while the fifth model is a new one. All five models allow for switching persistence of shocks, depending on the value and/or sign of recent returns. We consider the models for weekly data on 5 major stock markets. Our results indicate that all models improve upon the linear GARCH(1,1) model and that our new model sometimes yields favorable forecasting results. http://repub.eur.nl/res/pub/1407/ 1997-01-01T00:00:00Z The interest in business cycle asymmetry has been steadily increasing over the last fifteen years. Most research has focused on the different behaviour of macroeconomic variables during expansions and contractions, which by now is well documented. Recent evidence suggests that such a two-phase characterization of the business cycle might be too restrictive. In particular, it might be worthwhile to decompose the recovery phase in a high-growth phase (immediately following the trough of a cycle) and a subsequent moderate-growth phase. In this paper, the issue of multiple regimes is addressed using Smooth Transition AutoRegressive [STAR] models. A possible limitation of STAR models as they are currently used is that essentially they deal with only two regimes. We propose a generalization of the STAR model such that more than two regimes can be accommodated. It is demonstrated that the class of Multiple Regime STAR [MRSTAR] models can be obtained from the two-regime model in an elegant way. The main properties of the MRSTAR model and several issues which might be relevant for empirical specification are discussed in detail. In particular, a Lagrange Multiplier-type test is derived which can be used to determine the appropriate number of regimes. Application of the new model class to US real GNP and US unemployment rate provides evidence in favor of the existence of multiple business cycle phases. http://repub.eur.nl/res/pub/1408/ 1997-01-01T00:00:00Z A time series (t=921) of weekly survey data on vote intentions in the Netherlands for the period 1978-1995 shows that the percentage of undecided voters follows a cyclical pattern over the election calendar. The otherwise substantial percentage of undecided voters decreases sharply in weeks leading up to an election and gradually increases afterwards. This paper models the dynamics of this asymmetric electoral cycle using artificial neural networks, with the purpose of estimating when the undecided voters start making up their minds. We find that they begin to decide which party to vote for nine weeks before a first order national parliamentary election and one to four weeks before a second order election, depending on the type of election (European Parliament, Provincial States, City-councils). The effect of political campaigns and the implications for political analysis are discussed. http://repub.eur.nl/res/pub/1420/ 1997-01-01T00:00:00Z In this paper we test for (Generalized) AutoRegressive Conditional Heteroskedasticity [(G)ARCH] in daily and weekly data on 22 exchange rates and 13 stock market indices using the standard Lagrange Multiplier [LM] test for GARCH and a new LM test that is resistant to additive outliers. The data span two samples of 5 years ranging from 1986 to 1995. Our main result is that we find spurious GARCH in over 50% of the cases. Using Monte Carlo simulations, in which we evaluate our empirical method, we show that this general finding indeed appears to be due to outliers. We discuss some of the implications of our findings for empirical financial modeling. http://repub.eur.nl/res/pub/1421/ 1997-01-01T00:00:00Z In this paper we investigate empirical specification of smooth transition error correction models (STECMs). These models can be used to describe linear long-run relationships between nonstationary variables where adjustment towards equilibrium is nonlinear and can depend on exogenous variables. The various steps involved in specifying an appropriate model are discussed for a monthly bivariate interest rate series for The Netherlands. Using simulations we first establish that standard (linearity-based) cointegration tests can be used to examine joint long-run properties. Second, we apply various tests for nonlinearity to decide on an appropriate function for the adjustment of disequilibrium errors. When we estimate an STECM, we find indications that nonlinearity is due to only two observations. We investigate the relevance of these data points by applying robust tests for linearity and by considering less aggregated, i.e. weekly, data. We conclude with some suggestions for practitioners. http://repub.eur.nl/res/pub/2053/ 1997-01-01T00:00:00Z http://repub.eur.nl/res/pub/1382/ 1996-01-01T00:00:00Z Regime-switching models, like the smooth transition autoregressive (STAR) model are typically applied to time series of moderate length. Hence, the nonlinear features which these models intend to describe may be reflected in only a few observations. Conversely, neglected outliers in a linear time series of moderate length may incorrectly suggest STAR type nonlinearity. In this paper we propose outlier robust tests for STAR type nonlinearity. These tests are designed such that they have a better level and power behavior than standard nonrobust tests in situations with outliers. We formally derive local and global robustness properties of the new tests. Extensive Monte Carlo simulations show the practical usefulness of the robust tests. An application to several quarterly industrial production indices illustrates that apparent nonlinearity in time series sometimes seems due to only a small number of outliers. http://repub.eur.nl/res/pub/1395/ 1996-01-01T00:00:00Z In this paper we investigate the properties of the Lagrange Multiplier (LM) test for autoregressive conditional heteroskedasticity (ARCH) and generalized ARCH (GARCH) in the presence of additive outliers (AO's). We show analytically that both the asymptotic size and power are adversely affected if AO's are neglected: the test rejects the null hypothesis of homoskedasticity too often when it is in fact true, while the test has difficulty detecting genuine GARCH effects. Several Monte Carlo experiments show that these phenomena occur in small samples as well. We design and implement a robust test, which has better size and power properties than the conventional test in the presence of AO's. Applications to the French industrial production series and weekly returns of the Spanish peseta/US dollar exchange rate reveal that, sometimes, apparent GARCH effects may be due to only a small number of outliers and, conversely, that genuine GARCH effects can be masked by outliers. http://repub.eur.nl/res/pub/2110/ 1996-01-01T00:00:00Z In this paper we study the performance of the GARCH model and two of its non-linear modifications to forecast weekly stock market volatility. The models are the Quadratic GARCH (Engle and Ng, 1993) and the Glosten, Jagannathan and Runkle (1992) models which have been proposed to describe, for example, the often observed negative skewness in stock market indices. We find that the QGARCH model is best when the estimation sample does not contain extreme observations such as the 1987 stock market crash and that the GJR model cannot be recommended for forecasting.