http://repub.eur.nl/res/aut/4768/ List of Publications en http://repub.eur.nl/static-eur/img/logo.png http://repub.eur.nl http://repub.eur.nl/res/pub/37697/ 2012-06-01T00:00:00Z The average U.S. firm has less leverage than one would expect based on the trade-off between tax shields and bankruptcy costs. We focus on firms' financial flexibility and examine whether firms preserve debt capacity to reduce investment distortions in the future. We find that firms with high unused debt capacity invest more in future years than do firms with low unused debt capacity. Furthermore, firms that are reluctant to borrow in unconstrained periods are more likely to issue debt in periods in which access to capital markets is more constrained. http://repub.eur.nl/res/pub/22199/ 2010-12-01T00:00:00Z This study examines the liability hedging characteristics of both direct and indirect real estate with the advent of fair value accounting obligations for pension funds. We explicitly model pension obligations as being subject to interest and inflation risk to analyze the ability of real estate investments in hedging the fair value of pension liabilities and to quantify its role in an asset liability management (ALM) portfolio. We find that the portfolio composition differs depending on the definition of liability return. When liability returns solely follow actuarial changes, the mean-variance efficient portfolio allocations toward direct real estate and fixed income decrease compared to the asset-only optimization. When accounting for nominal liability obligations, real estate offers hedging benefits against interest rates for short holding periods but not for long-term institutional portfolios. The inclusion of inflation risk renders a limited role for direct real estate in an ALM portfolio, while indirect real estate obtains no allocation. Inflation is at the heart of the discrepancy between reported and predicted pension plan allocations. Once accounting for inflation, the projected allocations come close to reported ones. http://repub.eur.nl/res/pub/21278/ 2010-10-01T00:00:00Z This paper tests the static tradeoff theory against the pecking order theory. We focus on an important difference in prediction: the static tradeoff theory argues that a firm increases leverage until it reaches its target debt ratio, while the pecking order yields debt issuance until the debt capacity is reached. We find that for our sample of US firms the pecking order theory is a better descriptor of firms’ issue decisions than the static tradeoff theory. In contrast, when we focus on repurchase decisions we find that the static tradeoff theory is a stronger predictor of firms’ capital structure decisions. http://repub.eur.nl/res/pub/20317/ 2010-07-01T00:00:00Z This paper extends the basic pecking order model of Shyam-Sunder and Myers by separating the effects of financing surpluses, normal deficits, and large deficits. Using a panel of US firms over the period 1971-2005, we find that the estimated pecking order coefficient is highest for surpluses (0.90), lower for normal deficits (0.74), and lowest when firms have large financing deficits (0.09). These findings shed light on two empirical puzzles: 1) small firms, although having the highest potential for asymmetric information, do not behave according to the pecking order theory, and 2) the pecking order theory has lost explanatory power over time. We provide a solution to these puzzles by demonstrating that the frequency of large deficits is higher in smaller firms and increasing over time. We argue that our results are consistent with the debt capacity in the pecking order model. http://repub.eur.nl/res/pub/18574/ 2010-03-01T00:00:00Z Several Bayesian model combination schemes, including some novel approaches that simultaneously allow for parameter uncertainty, model uncertainty and robust time-varying model weights, are compared in terms of forecast accuracy and economic gains using financial and macroeconomic time series. The results indicate that the proposed time-varying model weight schemes outperform other combination schemes in terms of predictive and economic gains. In an empirical application using returns on the S&P 500 index, time-varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs. Another empirical example refers to forecasting US economic growth over the business cycle. It suggests that time-varying combination schemes may be very useful in business cycle analysis and forecasting, as these may provide an early indicator for recessions. http://repub.eur.nl/res/pub/16303/ 2009-07-16T00:00:00Z Several Bayesian model combination schemes, including some novel approaches that simultaneously allow for parameter uncertainty, model uncertainty and robust time varying model weights, are compared in terms of forecast accuracy and economic gains using financial and macroeconomic time series. The results indicate that the proposed time varying model weight schemes outperform other combination schemes in terms of predictive and economic gains. In an empirical application using returns on the S&P 500 index, time varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs. Another empirical example refers to forecasting US economic growth over the business cycle. It suggests that time varying combination schemes may be very useful in business cycle analysis and forecasting, as these may provide an early indicator for recessions. http://repub.eur.nl/res/pub/19481/ 2009-03-01T00:00:00Z We show that multifactor performance estimates for mutual funds suffer from systematic biases and argue that these biases are a result of miscalculating the factor premiums. Because the factor proxies are based on hypothetical stock portfolios and do not incorporate transaction costs, trade impact, and trading restrictions, the factor premiums are either over- or underestimated. We argue that factor proxies based on mutual fund returns rather than on stock returns provide better benchmarks to evaluate professional money managers. http://repub.eur.nl/res/pub/1833/ 2009-01-28T00:00:00Z In this paper we examine the usefulness of multivariate semi-parametric GARCH models for evaluating the Value-at-Risk (VaR) of a portfolio with arbitrary weights. We specify and estimate several alternative multivariate GARCH models for daily returns on the S&P 500 and Nasdaq indexes. Examining the within sample VaRs of a set of given portfolios shows that the semi-parametric model performs uniformly well, while parametric models in several cases have unacceptable failure rates. Interestingly, distributional assumptions appear to have a much larger impact on the performance of the VaR estimates than the particular parametric specification chosen for the GARCH equations. http://repub.eur.nl/res/pub/12676/ 2007-12-01T00:00:00Z A wide range of empirical biases hampers hedge fund databases. In this paper we focus upon survival-related biases and disentangle look-ahead biases due to self-selection of funds and due to fund termination. Self-selection arises because funds voluntarily report their information to data vendors and may decide to stop doing so. By extending existing methodology, we analyze persistence in hedge fund performance over the period 1994–2000, taking into account the above biases. The results show that look-ahead biases due to liquidation and self-selection enforce each other and may lead to overestimating expected returns by as much as 8% per year. Overall, the results are consistent with positive persistence in hedge fund returns at horizons of two and four quarters http://repub.eur.nl/res/pub/12672/ 2007-11-12T00:00:00Z In many countries there is a lack of genuine panel data where specific individuals or firms are followed over time. However, repeated cross-sectional surveys may be available, where a random sample is taken from the population at consecutive points in time. In this paper we discuss the identification and estimation of panel data models from repeated cross sections. In particular, attention will be paid to linear models with fixed individual effects, to models containing lagged dependent variables and to discrete choice models. http://repub.eur.nl/res/pub/12674/ 2007-10-12T00:00:00Z This study examines the liability hedging characteristic of both direct and indirect real estate, in the advent of fair value accounting obligations for pension funds. We explicitly model pension obligations as being subject to interest and inflation risk to analyze the ability of real estate investments in hedging the market value of pension liabilities and to quantify its role in an ALM portfolio. Based on a sample period of 1984-2006, direct and indirect real estate merit inclusion in an ALM portfolio because of their attractive risk-reward properties and its diversification potential, rather than its liability hedging abilities. http://repub.eur.nl/res/pub/12677/ 2007-08-01T00:00:00Z Copulas offer financial risk managers a powerful tool to model the dependence between the different elements of a portfolio and are preferable to the traditional, correlation-based approach. In this paper, we show the importance of selecting an accurate copula for risk management. We extend standard goodness-of-fit tests to copulas. Contrary to existing, indirect tests, these tests can be applied to any copula of any dimension and are based on a direct comparison of a given copula with observed data. For a portfolio consisting of stocks, bonds and real estate, these tests provide clear evidence in favor of the Student’s t copula, and reject both the correlation-based Gaussian copula and the extreme value-based Gumbel copula. In comparison with the Student’s t copula, we find that the Gaussian copula underestimates the probability of joint extreme downward movements, while the Gumbel copula overestimates this risk. Similarly we establish that the Gaussian copula is too optimistic on diversification benefits, while the Gumbel copula is too pessimistic. Moreover, these differences are significant. http://repub.eur.nl/res/pub/10451/ 2007-07-26T00:00:00Z Several frequentist and Bayesian model averaging schemes, including a new one that simultaneously allows for parameter uncertainty, model uncertainty and time varying model weights, are compared in terms of forecast accuracy over a set of simulation experiments. Artificial data are generated, characterized by low predictability, structural instability, and fat tails, which is typical for many financial-economic time series. Sensitivity of results with respect to misspecification of the number of included predictors and the number of included models is explored. Given the set up of our experiments, time varying model weight schemes outperform other averaging schemes in terms of predictive gains both when the correlation among individual forecasts is low and the underlying data generating process is subject to structural locations shifts. In an empirical application using returns on the S&P 500 index, time varying model weights provide improved forecasts with substantial economic gains in an investment strategy including transaction costs. http://repub.eur.nl/res/pub/12681/ 2007-03-01T00:00:00Z Using monthly return data of more than 6400 US equity mutual funds we investigate short-run performance persistence over the period 1984–2003. We sort funds into rank portfolios based on past performance, and evaluate the portfolios’ out-of-sample performance. To cope with short ranking periods, we employ an empirical Bayes approach to measure past performance more efficiently. Our main finding is that when funds are sorted into decile portfolios based on 12-month ranking periods, the top decile of funds earns a statistically significant, abnormal return of 0.26 percent per month. This effect persists beyond load fees, and is mainly concentrated in relatively young, small cap/growth funds. http://repub.eur.nl/res/pub/12670/ 2007-02-26T00:00:00Z This paper investigates the presence of spillover effects of marketing in mutual fund families. We find that funds with high marketing expenses generate spillovers, and enhance cash inflows to family members with low marketing expenses. In particular, low-marketing funds that are operated by a family with high marketing expenses have substantially larger inflows after positive returns than otherwise similar funds that are operated by a family with low marketing expenses, while they have smaller outflows after negative returns. One way to interpret the spillovers is that they are a by-product of individual fund marketing whereby the entire family is made more visible to investors. An alternative explanation of this observation is that funds with low marketing expenses are directly subsidized by family members with high marketing expenses. We develop and perform a set of tests to evaluate these two alternative hypotheses. The results of all tests support the subsidization hypothesis, and suggest that at least part of the spillovers can be attributed to favoritism. These results suggest that conflicts of interest between investors and fund families have been exacerbated by competition in the mutual fund industry. http://repub.eur.nl/res/pub/12673/ 2007-01-30T00:00:00Z This paper extends the basic pecking order model of Shyam-Sunder and Myers (1999) by separating the effects of financing surpluses, normal deficits, and large deficits. Using a broad cross-section of publicly traded firms for 1971 to 2005, we find that the estimated pecking order coefficient is highest for surpluses (0.90), lower for normal deficits (0.74), and lowest when firms have large financing deficits (0.09). These findings shed light on two empirical puzzles: first, small firms – although having the highest potential for asymmetric information – do not behave according to the pecking order theory, and second, the pecking order theory has lost explanatory power over time. We provide a solution to these puzzles by showing that the frequency of large deficits is higher in smaller firms and increasing over time. As a result, our findings support a pecking order theory that considers firms' debt capacities http://repub.eur.nl/res/pub/12667/ 2007-01-01T00:00:00Z We show that multifactor performance estimates for mutual funds suffer from systematic biases, and argue that these biases are a result of miscalculating the factor premiums. Because the factor proxies are based on hypothetical stock portfolios and do not incorporate transaction costs, trade impact, and trading restrictions, the factor premiums are either over- or underestimated. We argue that factor proxies based on mutual fund returns rather than stock returns provide better benchmarks to evaluate professional money managers. http://repub.eur.nl/res/pub/12669/ 2006-10-10T00:00:00Z Using monthly return data of more than 6,400 US equity mutual funds we investigate short-run performance persistence over the period 1984-2003. We sort funds into rank portfolios based on past performance, and evaluate the portfolios' out-of-sample performance. To cope with short ranking periods, we employ an empirical Bayes approach to measure past performance more efficiently. Our main finding is that when funds are sorted into decile portfolios based on 12-month ranking periods, the top decile of funds earns a statistically significant, abnormal return of 0.26 percent per month. This effect persists beyond load fees, and is mainly concentrated in relatively young, small cap/growth funds. http://repub.eur.nl/res/pub/12668/ 2006-09-11T00:00:00Z Copulas offer financial risk managers a powerful tool to model the dependence between the different elements of a portfolio and are preferable to the traditional, correlation-based approach. In this paper we show the importance of selecting an accurate copula for risk management. We extend standard goodness-of-fit tests to copulas. Contrary to existing, indirect tests, these tests can be applied to any copula of any dimension and are based on a direct comparison of a given copula with observed data. For a portfolio consisting of stocks, bonds and real estate, these tests provide clear evidence in favor of the \\studt copula, and reject both the correlation-based Gaussian copula and the extreme value-based Gumbel copula. In comparison with the \\studt copula, we find that the Gaussian copula underestimates the probability of joint extreme downward movements, while the Gumbel copula overestimates this risk. Similarly we establish that the Gaussian copula is too optimistic on diversification benefits, while the Gumbel copula is too pessimistic. Moreover, these differences are significant. http://repub.eur.nl/res/pub/12613/ 2006-08-01T00:00:00Z Systemic crises can have grave consequences for investors in international equity markets, because they cause the risk-return trade-off to deteriorate severely for a longer period. We propose a novel approach to include the possibility of systemic crises in asset allocation decisions. By combining regime switching models with Merton [Merton, R.C., 1969. Lifetime portfolio selection under uncertainty: The continuous time case. Review of Economics and Statistics 51, 247–257]-style portfolio construction, our approach captures persistence of crises much better than existing models. Our analysis shows that incorporating systemic crises greatly affects asset allocation decisions, while the costs of ignoring them is substantial. For an expected utility maximizing US investor, who can invest globally these costs range from 1.13% per year of his initial wealth when he has no prior information on the likelihood of a crisis, to over 3% per month if a crisis occurs with almost certainty. If a crisis is taken into account, the investor allocates less to risky assets, and particularly less to the crisis prone emerging markets. http://repub.eur.nl/res/pub/7875/ 2006-07-14T00:00:00Z Believers in the law of small numbers tend to overinfer the outcome of a random process after a small series of observations. They believe that small samples replicate the probability distribution properties of the population. We provide empirical evidence indicating that investors are mistakenly driven by this psychological bias when hiring or firing a fund manager after a successful (or losing) performance streak. Using quarterly data between 1994 and 2000 of 752 hedge funds, we analyze actual money flows into and out of hedge funds and their relationship with the length of the streak. We first show that persistence patterns have a predictive ability of future relative performance of a manager: the longer the winner streak, the larger the probability for a fund to remain a winner. Investors, in turn, appear to be aware of quality dispersion across managers and respond by following a momentum strategy: the longer the winning (losing) streak, the more likely they will invest in (divest from) that fund. Yet, we find that investors place excessive weight in the managers’ track record as a criterion for decision. Our model shows that the length of the streak has an economically and statistically significant impact on money flows beyond rationally expected performance, which confirms a “hot-hand” bias driving to a large extent momentum investing. Apparently, even sophisticated investors exhibit psychological biases that may have adverse effects on their wealth. http://repub.eur.nl/res/pub/7096/ 2005-11-18T00:00:00Z We explore the flow-performance interrelation by explicitly separating the investment and divestment decisions of hedge fund investors. The results show that different determinants and evaluation horizons underlie both decisions. While money inflows are sensitive to past long-run performance, outflows exhibit an immediate and sustained response to past performance in the short run. As a consequence, the shape of the flow-performance relation differs depending on the time horizon being analyzed. We find a weaker flow-performance relation for winning funds at quarterly horizons compared to annual horizons, which may explain why quarterly persistence in hedge fund performance is not competed away. Indeed, we also find evidence that most investors are unable to exploit the persistence of the winners. Conversely, investors are fast and successful in deallocating from the persistent losers, ensuring a disciplining mechanism for lowquality funds. Further, our findings do not support the existence of smart money. http://repub.eur.nl/res/pub/12671/ 2005-08-07T00:00:00Z Systemic crises can have grave consequences for investors in international equity markets, because it causes the risk-return trade-off to deteriorate severely for a longer period. In this paper we propose a novel approach to include the possibility of systemic crises in asset allocation decisions. By combining regime switching models with Merton (1969)-style portfolio construction, our approach captures persistence of crises much better than existing models. Our analysis shows that incorporating systemic crises has a large impact on asset allocation decisions, while the costs of ignoring such crises are substantial. For an expected utility maximizing US investor, who can invest globally these costs range from 1.13% per year of his initial wealth when he has no prior information on the likelihood of a crisis, to over 3% per month if a crisis occurs with almost certainty. If a crisis is taken into account, the investor allocates less to risky assets, and particularly less to emerging markets, being most prone to a crisis. An investor facing short selling constraints withdraws completely from equity markets if the likelihood of a crisis increases. http://repub.eur.nl/res/pub/12615/ 2005-07-01T00:00:00Z An important feature of panel data is that it allows the estimation of parameters characterizing dynamics from individual level data. Several authors argue that such parameters can also be identified from repeated cross-section data and present estimators to do so. This paper reviews the identification conditions underlying these estimators. As grouping data to obtain a pseudo-panel is an application of instrumental variables (IV), identification requires that standard IV conditions are met. This paper explicitly discusses the implications of these conditions for empirical analyses. We also propose a computationally attractive IV estimator that is consistent under essentially the same conditions as existing estimators. While a Monte Carlo study indicates that this estimator may work well under relatively weak conditions, these conditions are not trivially satisfied in applied work. Accordingly, a key conclusion of the paper is that these estimators cannot be implemented under general conditions. http://repub.eur.nl/res/pub/12614/ 2005-01-01T00:00:00Z We analyze the performance persistence in hedge funds taking into account look-ahead bias (multi-period sampling bias). We model liquidation of hedge funds by analyzing how it depends upon historical performance. Next, we use a weighting procedure that eliminates look-ahead bias in measures for performance persistence. In contrast to earlier results for mutual funds, the impact of look-ahead bias is exacerbated for hedge funds due to their greater level of total risk. At the four-quarter horizon, look-ahead bias can be as much as 3.8%, depending upon the decile of the distribution. We find positive persistence in hedge fund quarterly returns after correcting for investment style. The empirical pattern at the annual level is also consistent with positive persistence, but its statistical significance is weak. http://repub.eur.nl/res/pub/1822/ 2004-12-10T00:00:00Z A wide range of empirical biases hampers hedge fund databases. In this paper we focus upon survival-related biases and disentangle look-ahead biases due to self-selection of funds and due to fund termination. Self-selection arises because funds voluntarily report their information to data vendors and may decide to stop doing so. By extending existing methodology, we analyze persistence in hedge fund performance over the period 1994-2000, taking into account the above biases. The results show that look-ahead biases due to liquidation and self-selection enforce each other and may lead to overestimating expected returns by as much as 8% per year. Overall, the results are consistent with positive persistence in hedge fund returns at horizons of two and four quarters. http://repub.eur.nl/res/pub/12628/ 2004-12-01T00:00:00Z This paper develops a novel approach to simultaneously test for market timing in stock index returns and volatility. The tests are based on the estimation of a system of regression equations with indicator variables and provide detailed information about the statistical significance of alternative market timing components. http://repub.eur.nl/res/pub/12630/ 2004-09-01T00:00:00Z This paper investigates the question whether individual stock momentum in Europe is subsumed by country or industry momentum. We introduce a portfolio-based regression approach, which directly allows to test hypotheses about the existence and relative importance of multiple effects (e.g., momentum, value, and size), even when only a moderate number of stocks are available. Our results suggest that the positive expected excess returns of momentum strategies in European stock markets are primarily driven by individual stock effects, while industry momentum plays a less important role and country momentum is even weaker. These results are robust to the inclusion of value and size effects. http://repub.eur.nl/res/pub/12675/ 2004-07-05T00:00:00Z This paper develops a novel approach to simultaneously test for market timing in stock index returns and volatility. The tests are based on the estimation of a system of regression equations with indicator variables and provide detailed information about the statistical significance of alternative market timing components. http://repub.eur.nl/res/pub/1333/ 2004-06-23T00:00:00Z This paper investigates three capital structure decisions – leverage, debt maturity and the source of debt – in a simultaneous setting. Moreover, we investigate whether these choices are influenced by the involvement of banks in a firm. Our results based on a panel of Dutch firms show that bank relationships, measured by interlocking board memberships and equity ownership, have a significant impact on the relations among the three capital structure choices. First, less bank involvement strengthens the positive impact of leverage on maturity. This is consistent with the liquidity risk theory, because involved banks help firms to mitigate liquidity risk. Second, bank debt negatively effects leverage in firms with bank interlocks, while this relation is absent in firms without such bank involvement. This result suggests that banks maximize the value of their loans by reducing overall leverage. Third, we find a strong trade-off between bank debt and maturity, which is independent of the degree of bank involvement. http://repub.eur.nl/res/pub/1270/ 2004-04-17T00:00:00Z Systemic crises can largely affect asset allocations due to the rapid deterioration of the risk-return trade-off. We investigate the effects of systemic crises, interpreted as global simultaneous shocks to financial markets, by introducing an investor adopting a crisis ignorant or crisis conscious strategy. Including the possibility of a systemic crisis is a substantial improvement. Investments in risky assets fall, while allocations to countries less sensitive to a crisis grow relatively. An increasing probability of a crisis exacerbates these effects. The certainty equivalent costs of ignoring systemic crises are large, ranging from 0.65% per year unconditionally, to over 5% per month conditionally on a high probability for the occurrence of a crisis. http://repub.eur.nl/res/pub/12611/ 2004-01-01T00:00:00Z A Guide to Modern Econometrics is a new textbook published by John Wiley and Sons. It covers a wide range of topics in applied econometrics in a concise and intuitive way. Some distinctive features: Emphasis on empirical relevance and intuition, paying attention to the links between alternative approaches. Limited use of matrix algebra. Coverage of many modern topics from time-series, cross-section and panel data econometrics. Concisely and carefully written, so that the reader does not get lost in the details. Full length empirical illustrations are provided throughout, typically taken from the modern economics literature and using full-size data sets. Empirical illustrations taken from finance, labour economics, environmental economics, monetary economics, international economics and many more. Exercises added to all chapters, with a focus on intuition and interpretation of results. Several exercises involve the use of actual data. Data sets used for illustrations and exercises are available from the internet. http://repub.eur.nl/res/pub/12629/ 2004-01-01T00:00:00Z In this paper, we analyze the economic value of predicting stock index returns as well as volatility. On the basis of simple linear models, estimated recursively, we produce out-of-sample forecasts for the return on the S&P 500 index and its volatility. Using monthly data, we examine the economic value of a number of alternative trading strategies over the period 1970-2001. It appears easier to forecast returns at times when volatility is high. For a mean-variance investor, this predictability is economically profitable, even if short sales are not allowed and transaction costs are quite large. The economic value of trading strategies that employ market timing in returns and volatility exceeds that of strategies that only employ timing in returns. Most of the profitability of the dynamic strategies, however, is located in the first half of our sample period http://repub.eur.nl/res/pub/978/ 2003-10-20T00:00:00Z We decompose the conditional expected mutual fund return in five parts. Two parts, selectivity and expert market timing, can be attributed to manager skill, and three to variation in market exposure that can be achieved by private investors as well. The dynamic model that we use to estimate the relative importance of the components in the decomposition is a generalization of the performance evaluation models by Lockwood and Kadiyala (1988) and Ferson and Schadt (1996). We find that the restrictions imposed in existing models may lead to different inferences about manager selectivity and timing skill. The results from our sample of 78 asset allocation mutual funds indicate that several funds exhibit significant expert market timing, but for most funds variation in market exposures does not yield any economically significant return. Funds with high turnover and expense ratios are associated with managers with better skills. http://repub.eur.nl/res/pub/923/ 2003-09-24T00:00:00Z In this study we propose the use of the Student's t dependence function to model dependence between asset returns when conducting stress tests. To properly include stress testing in a risk management system, it is important to have accurate information about the (joint) probabilities of extreme outcomes. Consequently, a model for the behavior of risk factors is necessary, specifying the marginal distributions and their dependence. Traditionally, dependence is described by a correlation matrix, implying the use of the dependence function inherent in the multivariate normal (Gaussian) distribution. Recent studies have cast serious doubt on the appropriateness of the Gaussian dependence function to model dependence between extreme negative returns. The student's t dependence function provides an attractive alternative. In this paper, we introduce four tests to analyze the empirical fit of both dependence functions. The empirical results indicate that probabilities assigned to stress tests are largely influenced by the choice of dependence function. The statistical tests reject the Gaussian dependence function, but do not reject the Student's t dependence function. http://repub.eur.nl/res/pub/255/ 2002-11-19T00:00:00Z Hedge funds databases are typically subject to high attrition rates because of fund termination and self-selection. Even when all funds are included up to their last available return, one cannot prevent that ex post conditioning biases a.ect standard estimates of performance persistence. In this paper we analyze the persistence in the performance of U.S. hedge funds taking into account look-ahead bias (multi-period sampling bias). To do so, we model attrition of hedge funds and analyze how it depends upon historical performance. Next, we use a weighting procedure that eliminates look-ahead bias in measures for performance persistence. The results show that the impact of look-ahead bias is quite severe, even though positive and negative survival-related biases are sometimes suggested to cancel out. At horizons of one and four quarters, we find clear evidence of positive persistence in hedge fund returns, also after correcting for investment style. At the two-year horizon, past winning funds tend to perform poorly in the future. http://repub.eur.nl/res/pub/246/ 2002-10-28T00:00:00Z The driving force behind the well-documented medium term momentum effect in stock returns is subject of much debate. Empirical papers that aim to find the determinants of this return continuation, seem to be almost exclusively restricted to US stock markets. Consequently, regional effects have received little attention in these analyses. This paper contributes to the discussion by investigating the presence of country and industry momentum in Europe and addressing the question whether individual stock momentum is subsumed by country or industry momentum.We examine these issues by introducing a portfolio-based regression approach, which allows to test hypotheses about the existence and relative importance of multiple effects using standard statistical techniques. While the traditional sorting techniques are not suited to disentangle a multitude of possibly interrelated effects (e.g. momentum, value, and size), our method can be used even when only a moderate number of stocks are available. Our results suggest that the positive expected excess returns of momentum strategies in European stock markets are primarily driven by individual stocks effects, while industry momentum plays a less important role and country momentum is even weaker. These results are robust to the inclusion of value and size effects. http://repub.eur.nl/res/pub/343/ 2002-06-21T00:00:00Z Rede, in verkorte vorm uitgesproken bij de aanvaarding van het ambt van hoogleraar Ondernemingsfinanciering aan de Faculteit der Bedrijfskunde en de Faculteit der Economische Wetenschappen van de Erasmus Universiteit Rotterdam op vrijdag 21 juni 2002 http://repub.eur.nl/res/pub/582/ 2002-02-13T00:00:00Z An important feature of panel data is that it allows the estimation of parameters characterizing dynamics from individual level data. Several authors argue that such parameters can also be identified from repeated cross-section data and present estimators to do so. This paper reviews the identification conditions underlying these estimators. As grouping data to obtain a pseudo-panel is an application of instrumental variables (IV), identification requires that standard IV conditions are met. This paper explicitly discuss the implications of these conditions for empirical analyses. We also propose a computationally attractive instrumental variables estimator that is consistent under a relatively weak set of conditions. A Monte Carlo study indicates that this estimator may work well in practice. http://repub.eur.nl/res/pub/133/ 2001-12-10T00:00:00Z In this paper, we analyze the economic value of predicting stock index returns as well as volatility. On the basis of simple linear models, estimated recursively, we produce genuine out-of-sample forecasts for the return on the S&P 500 index and its volatility. Using monthly data from 1954 to 2001, we test the statistical significance of return and volatility predictability and examine the economic value of a number of alternative trading strategies. While we find strong evidence for market timing in both returns and volatility, the success of market timing and volatility timing varies considerably over the sample period. Further, it appears easier to forecast returns at times when volatility is high. For a mean-variance investor, this predictability is economically profitable, even if short sales are not allowed and transaction costs are quite large. The economic value of trading strategies that employ market timing in returns and volatility exceeds that of strategies that only employ timing in returns. http://repub.eur.nl/res/pub/12631/ 2001-09-01T00:00:00Z Performance persistence studies typically suffer from ex-post conditioning biases. As stressed by Carhart [Carhart, M.M., 1997. Mutual Fund Survivorship, Working Paper, Marshall School of Business, U.S.C.] and Carpenter and Lynch [J. Financ. Econ. 54 (1999) 337.], standard methods of analysis on a survivorship free sample are subject to look-ahead biases. In this paper, we show how one can easily correct for look-ahead bias using weights based on probit regressions. First, we model how survival probabilities depend upon historical returns, fund age and aggregate economy-wide shocks, using two samples of US based ‘income’ and ‘growth’ funds. Subsequently, we employ a Monte Carlo study to analyze the size and shape of the look-ahead bias in performance persistence that arise when a survivorship free sample is used with standard techniques. In particular, we show that look-ahead bias induces a spurious U-shaped pattern in performance persistence. Finally, we demonstrate how a weighting procedure based upon probit regressions can be used to correct for this bias. In this way, we obtain look-ahead bias-corrected estimates of abnormal performance relative to a one-factor and the Carhart [J. Finan. 52 (1997) 57.] four-factor model, as well as its persistence. The results suggest that in this sample, look-ahead bias is of minor importance and does not seriously affect estimates of persistence. Our bias-corrected results closely correspond to the findings of Carhart [J. Finan. 52 (1997) 57.], implying that there is no evidence on a risk-adjusted basis for persistence in performance. http://repub.eur.nl/res/pub/12632/ 2000-01-01T00:00:00Z This paper analyzes the properties of a number of estimators that can be used to estimate short-run persistence in mutual fund returns. When data for different funds are pooled, it is advisable to correct for cross-sectional differences in expected returns. However, these adjustments may induce biases in the estimated persistence coefficients and thus lead to spurious persistence. Theoretical derivations, combined with a Monte Carlo study, show that these biases cannot be neglected for the samples that are typically used in applied work. We also estimate the short-run persistence in two samples of U.S. open-end mutual funds using quarterly returns for 1987-1994. An important conclusion is that the results are quite sensitive to the estimation method that is employed. http://repub.eur.nl/res/pub/12635/ 1999-01-01T00:00:00Z The relationship between two alternative approaches, instrumental variables and control function procedures, for estimating the impact of endogenous treatment effects are examined. Although it is well known that the two approaches generate comparable estimates, the relationship between the estimators and their accompanying endogeneity tests appears not to be well understood. It is shown that the two procedures are closely related. The implications of the two procedures for the underlying economic sorting behavior are also examined. http://repub.eur.nl/res/pub/12640/ 1999-01-01T00:00:00Z This paper employs the rank-order instrumental variable (IV) procedure of Vella and Verbeek [Vella, F., Verbeek, M., 1997. Using rank order as an instrumental variable: an application to the return to schooling, CES Discussion Paper 97.10, K.U. Leuven.] to estimate the returns to education for Australian youth. The attraction of this approach is that it can account for the endogeneity of schooling in the wage equation via the use of instrumental variables without the use of exclusion restrictions. We find, after accounting for the endogeneity of schooling, that an additional year of schooling is associated with an increase in wages of approximately 8%. Furthermore, we find that the rank-order IV approach is able to identify the presence of endogeneity in this particular empirical example. However, despite this, the adjusted estimate of how schooling affects wage is close to the ordinary least squares (OLS) estimate. http://repub.eur.nl/res/pub/12641/ 1999-01-01T00:00:00Z In intertemporal asset pricing models, transaction costs are usually neglected. In this paper we explicitly incorporate transaction costs in these models and analyze to what extent this extension is helpful in explaining the cross-section of expected returns. An empirical analysis using CRSP data on size-based portfolios examines the role of the transaction costs and shows that incorporating such costs in the consumption-based model with power utility does not yield very satisfactory results. However, the introduction of habit persistence substantially improves the model. We find rather strong evidence of habit persistence in monthly consumption data. The plots of the models' pricing errors indicate that an intertemporal asset pricing model with transaction costs and habit persistence explains the cross-sectional variation in the portfolio returns quite accurately. http://repub.eur.nl/res/pub/12642/ 1999-01-01T00:00:00Z This paper presents some two-step estimators for a wide range of parametric panel data models with censored endogenous variables and sample selection bias. Our approach is to derive estimates of the unobserved heterogeneity responsible for the endogeneity/selection bias to include as additional explanatory variables in the primary equation. These are obtained through a decomposition of the reduced form residuals. The panel nature of the data allows adjustment, and testing, for two forms of endogeneity and/or sample selection bias. Furthermore, it incorporates roles for dynamics and state dependence in the reduced form. Finally, we provide an empirical illustration which features our procedure and highlights the ability to test several of the underlying assumptions. http://repub.eur.nl/res/pub/12643/ 1998-01-01T00:00:00Z We estimate the union premium for young men over a period of declining unionization (1980-87) through a procedure which identifies the alternative sources of the endogeneity of union status. While we estimate the average increase in wages resulting from union employment to be in excess of 20% we find that the return to unobserved heterogeneity operating through union status is substantial and that the union premium is highly variable. We also find that the premium is sensitive to the form of sorting allowed in estimation. Moreover, the data are consistent with comparative advantage sorting. Our results suggest that the unobserved heterogeneity which positively contributes to the likelihood of union membership is associated with higher wages. We are unable, however, to determine whether this is due to the ability of these workers to extract monopoly rents or whether it reflects the more demanding hiring standards of employers faced by union wages. http://repub.eur.nl/res/pub/12646/ 1995-01-01T00:00:00Z In a panel data model with fixed individual effects, a number of alternative transformations are available to eliminate these effects such that the slope parameters can be estimated from ordinary least squares on transformed data. In this note we show that each transformation leads to algebraically the same estimator if the transformed data are used efficiently (i.e. if GLS is applied). If OLS is used, however, differences may occur and the routinely computed variances, even after degrees of freedom correction, are incorrect. In addition, it may matter whether “redundant” observations are used or not. http://repub.eur.nl/res/pub/12647/ 1993-01-01T00:00:00Z In this paper it is argued that maximizing the complete data (log) likelihood function with respect to the missing data and the unknown parameters will not improve the efficiency of the estimators but may affect consistency instead. If no auxiliary relations are available or additional assumptions are made, the maximum likelihood estimator based on the observed data is (asymptotically) the most efficient estimator. http://repub.eur.nl/res/pub/12648/ 1993-01-01T00:00:00Z If panel data are not available but repeated cross-sections are, the parameters in a regression model with fixed individual effects can be estimated consistently using the cohort approach proposed by Deaton (1985). In this paper we show that Deaton's estimator is inconsistent if the number of time periods is small, even if the number of cohorts tends to infinity. Moreover, we propose an alternative estimator which does not suffer from a bias due to a small number of sampling periods and we introduce a new class of estimators, containing both estimators mentioned above. We discuss minimum mean squared error estimation within this class. Our results show that it may be optimal to eliminate only part of the measurement error in the cohort averages, since the implied bias is offset by a smaller variance. http://repub.eur.nl/res/pub/12651/ 1992-03-01T00:00:00Z If repeated observations on the same individuals are not available it is not possible to capture unobserved individual characteristics in a linear model by using the standard fixed effects estimator. If large numbers of observations are available in each period one can use cohorts of individuals with common characteristics to achieve the same goal, as shown by Deaton (1985). It is tempting to analyze the observations on cohort averages as if they are observations on individuals which are observed in consecutive time periods. In this paper we analyze under which conditions this is a valid approach. Moreover, we consider the impact of the construction of the cohorts on the bias in the standard fixed effects estimator. Our results show that the effects of ignoring the fact that only a synthetic panel is available will be small if the cohort sizes are sufficiently large (100, 200 individuals) and if the true means within each cohort exhibit sufficient time variation. http://repub.eur.nl/res/pub/12649/ 1992-01-01T00:00:00Z In this note we consider the optimal experimental design for cases in which the data may consist of both cross-sectional and panel observations. Our results generalize those of Aigner and Balestra (1988) on the choice of controls and pre-experimental observationsby avoiding the assumption that the marginal costs of additional observations on the same unit are negligible. We derive conditions under which the panel design considered by Aigner and Balestra is nevertheless optimal as well as conditions under which reinterviews are inefficient. http://repub.eur.nl/res/pub/12650/ 1992-01-01T00:00:00Z Discusses several tests to check for the presence of selectivity bias in estimators based on panel data. Introduction; Selectivity bias in the fixed and random effects estimators; Simple tests for selectivity bias; Specification of the response mechanism and the LM test for selectivity bias; Numerical results on the pseudo true values of the RE and FE estimators; More. http://repub.eur.nl/res/pub/12652/ 1992-01-01T00:00:00Z If missing observations in a panel data set are not missing at random, many widely applied estimators may be inconsistent. In this paper we examine empirically several ways to reveal the nature and severity of the selectivity problem due to nonresponse, as well as a number of methods to estimate the resulting models. Using a life cycle consumption function and data from the Expenditure Index Panel from the Netherlands, we discuss simple procedures that can be used to assess whether observations are missing at random, and we consider more complicated estimation procedures that can be used to obtain consistent or efficient estimates in case of selectivity of attrition bias. Finally, some attention is paid to the differences in identification, consistency, and efficiency between inferences from a single wave of the panel, a balanced sub-panel, and an unbalanced panel. http://repub.eur.nl/res/pub/12661/ 1991-01-01T00:00:00Z In this paper we consider the relative efficiency of rotating-panel designs in analysis-of-variance models. Throughout we assume that the parameter of interest is a linear combination of period means in the analysis-of-variance model. Results from spectral theory are used to obtain manageable expressions for the variance of the BLUE of this parameter. Relative efficiencies of the BLUE for rotating panels with different rotation periods are presented, e.g., for the period means themselves, of differences, or of averages of means. Moreover we present bounds on the relative efficiency which are valid irrespective of the parameter of interest. The analysis shows that the gains from choosing an optimal rotation design can be quite substantial, even if the cost of a reinterview equals the cost of a first observation. In many cases either the smallest or the highest possible rotation period is optimal. The analysis is illustrated with an empirical example concerning monthly consumer expenditures on food and clothing. http://repub.eur.nl/res/pub/12662/ 1990-01-01T00:00:00Z In case of sample selectivity the maximum likelihood estimator of the parameters in a model with fixed effects will not be consistent when the number of time periods is small. In this paper, we present a transformation to eliminate the fixed individual effects and show that the corresponding marginal maximum likelihood estimator is computationally feasible and can be used to estimate the remaining parameters consistently even if number of time periods is finite. http://repub.eur.nl/res/pub/12663/ 1990-01-01T00:00:00Z In this paper we consider the estimation of time-dependent parameters in linear models from panel data, cross-sections, or both. We determine the fraction of individuals that should be reinterviewed each period in order to minimize the variance of the most efficient estimator of linear combinations of the parameters. Moreover we derive simple sufficient conditions for the optimal fraction to be zero or one, respectively.