T.E. Nijman (Theo)
http://repub.eur.nl/ppl/5106/
List of Publicationsenhttp://repub.eur.nl/eur_logo.png
http://repub.eur.nl/
RePub, Erasmus University RepositoryAn Anatomy of Commodity Futures Risk Premia
http://repub.eur.nl/pub/76257/
Sat, 01 Feb 2014 00:00:01 GMT<div>M. Szymanowska</div><div>F. de Roon</div><div>T.E. Nijman</div><div>R. van den Goorbergh</div>
We identify two types of risk premia in commodity futures returns: spot premia related to the risk in the underlying commodity, and term premia related to changes in the basis. Sorting on forecasting variables such as the futures basis, return momentum, volatility, inflation, hedging pressure, and liquidity results in sizable spot premia between 5% and 14% per annum and term premia between 1% and 3% per annum. We show that a single factor, the high-minus-low portfolio from basis sorts, explains the cross-section of spot premia. Two additional basis factors are needed to explain the term premia.Do Countries or Industries Explain Momentum in Europe?
http://repub.eur.nl/pub/12630/
Wed, 01 Sep 2004 00:00:01 GMT<div>T.E. Nijman</div><div>L.A.P. Swinkels</div><div>M.J.C.M. Verbeek</div>
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.Do Countries or Industries Explain Momentum in Europe?
http://repub.eur.nl/pub/246/
Mon, 28 Oct 2002 00:00:01 GMT<div>T.E. Nijman</div><div>L.A.P. Swinkels</div><div>M.J.C.M. Verbeek</div>
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.Eliminating Look-Ahead Bias in Evaluating Persistence in Mutual Fund Performance
http://repub.eur.nl/pub/12631/
Sat, 01 Sep 2001 00:00:01 GMT<div>J.R. ter Horst</div><div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div>
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.Evaluating Style Analysis
http://repub.eur.nl/pub/25/
Thu, 25 May 2000 00:00:01 GMT<div>F.A. de Roon</div><div>T.E. Nijman</div><div>B.J.M. Werker</div>
In this paper we evaluate applications of (return based) style analysis. The portfolio and positivity constraints imposed by style analysis are useful in constructing mimicking portfolios without short positions. Such mimicking portfolios can be used e.g. to construct efficient portfolios of mutual funds with desired factor loadings if the factor loadings in the underlying factor model are positively weighted portfolios. Under these conditions style analysis may also be used to determine a benchmark portfolio for performance measurement. Attribution of the returns on portfolios of which the actual composition is unobserved to specific asset classes on the basis of return based style analysis is attractive if moreover there are no additional cross exposures between the asset classes and if fund managers hold securities that on average have a beta of one relative to their own asset class. If such restrictions are not met, and in particular if the factor loadings do not generate a positively weighted portfolio, the restrictions inherent in return based style analysis distort the outcomes of standard regression approaches rather than that the analysis is improved. The size of the distortions is illustrated by considering empirical results on style analysis of US mutual funds.Currency Hedging for International Stock Portfolios
http://repub.eur.nl/pub/24/
Wed, 24 May 2000 00:00:01 GMT<div>F.A. de Roon</div><div>T.E. Nijman</div><div>B.J.M. Werker</div>
This paper tests whether hedging currency risk improves the performance of international stock portfolios. We use a generalized performance measure which allows for investor-dependencies such as different utility functions and the presence of nontraded risks. In addition we show that an auxiliary regression, similar to the Jensen regression, provides a wealth of information about the optimal portfolio holdings for investors for the non mean-variance case. This is analogous to the information provided by the Jensen regression about optimal portfolio holdings for the mean-variance case. Our empirical results show that static hedging with currency forwards does not lead to improvements in portfolio performance for a US investor that holds a stock portfolio from the G5 countries. On the other hand, hedges that are conditional on the current interest rate spread do lead to significant performance improvements. Also, when an investor has a substantial exogenous exposure to one of the currencies, currency hedging clearly improves his portfolio performance. While these results hold for investors with power utility as well as with mean-variance utility functions, the optimal hedge ratios for these investors are different.Minimum MSE estimation of a regression model with fixed effects from a series of cross-sections
http://repub.eur.nl/pub/12648/
Fri, 01 Jan 1993 00:00:01 GMT<div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div>
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.Can cohort data be treated as genuine panel data?
http://repub.eur.nl/pub/12651/
Sun, 01 Mar 1992 00:00:01 GMT<div>M.J.C.M. Verbeek</div><div>T.E. Nijman</div>
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.The optimal choice of controls and pre-experimental observations
http://repub.eur.nl/pub/12649/
Wed, 01 Jan 1992 00:00:01 GMT<div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div>
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.Testing for selectivity bias in panel data models
http://repub.eur.nl/pub/12650/
Wed, 01 Jan 1992 00:00:01 GMT<div>M.J.C.M. Verbeek</div><div>T.E. Nijman</div>
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.Nonresponse in panel data: The impact on estimates of a life cycle consumption function
http://repub.eur.nl/pub/12652/
Wed, 01 Jan 1992 00:00:01 GMT<div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div>
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.The efficiency of rotating-panel designs in an analysis-of-variance model
http://repub.eur.nl/pub/12661/
Tue, 01 Jan 1991 00:00:01 GMT<div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div><div>A.H.O. van Soest</div>
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.Estimation of time-dependent parameters in linear models using cross-sections, panels, or both
http://repub.eur.nl/pub/12663/
Mon, 01 Jan 1990 00:00:01 GMT<div>T.E. Nijman</div><div>M.J.C.M. Verbeek</div>
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.