Search

Risk related to return higher moments is usually perceived as substantial in the hedge fund industry. In this article, we relate the return co-skewness and co-kurtosis of hedge fund strategies to macroeconomic and financial shocks. We find that hedge funds’ return higher moments are quite sensitive to these shocks, and especially during the subprime crisis. Our robust IV-GMM estimation procedure reveals that output shocks are more important for explaining the co-kurtosis of most hedge fund strategies than shocks related to jumps in the VIX, while we find the opposite for co-skewness. This suggests that hedge funds better monitor their co-kurtosis. Our VAR experiments show that the bulk of the reaction of higher moment risk to shocks occurs during recessions. Furthermore, some directional strategies—like short-sellers and futures—successfully track shocks and deliver higher returns during periods of crisis, benefiting from market volatility analogously to lookback straddles. Our results also suggest that illiquidity factors are subject to endogeneity and may be viewed as indicators of return smoothing, especially during crises. Robustness checks show that the practice of return smoothing may lead to a serious underestimation of hedge fund higher moment systematic risk, and particularly fat-tail risk.

We study how downside risk taken by hedge fund strategies responds to macroeconomic and financial shocks. Using new empirical measures of systematic tail risk, we find that the impulse response functions of strategies’ multi-moment risk display an asymmetric behavior, being much more significant and nonlinear during the subprime crisis than during the recent expansion period. Our results indicate that managers of hedge fund strategies seek to monitor their tail risk during crises, in the sense that their portfolios seem to behave as puts, which are kurtosis reducers in times of turmoil. However, while some strategies (e.g., futures) succeed quite well in controlling their tail risk during crises, others (e.g., equity market neutral) have difficulties to do so. This is an important result for institutional investors who think that hedge funds are “hedged” and thus immunized against adverse macroeconomic shocks—especially market volatility shocks.

International audience; This multidisciplinary article attempts to bridge the gap between policy-driven, agent-driven, psychological and market variables that may be factors in financial crises and may have been involved in the 2007–2009 Global Financial Crisis (GFC), which we use as an example. We conducted a quantitative study performed on 202 participants, who acted as proxies for consumers during that period. The results of this exploratory research tend to explain how transactions between eager borrowers/homebuyers and avid lenders developed in the context of a predatory broth, which had deceit at its center. This understanding can help regulators to canvas the use of financial tools and advertising techniques to minimize the negative impact on the economy.

The Fama-French (FF) five-factor model is cast into a dynamic setting to capture the impact of illiquidity over the phases of the business cycle on the returns of the passive FF twelve sector portfolios. We use two dynamic approaches, Kalman filtering and a recursive/rolling robust instrumental variables (IV) algorithm cast into a GMM framework, to determine time-varying alpha and beta estimates. Our principal result is that the Kalman filter approach supports the hypothesis that illiquidity is an important risk factor in a dynamic context. However, the only factor found to matter in the dynamic GMM approach is the market risk premium. Nevertheless, illiquidity may be prescient with respect to financial crises. Keywords: Illiquidity, Fama-French five-factor model, Kalman filter, Robust IV algorithm, JEL Classification: C58, G12

We investigate how macroeconomic and financial uncertainty impacts the behavior of hedge fund strategy higher moments—i.e., co-skewness and co-kurtosis—and their respective cross-sectional dispersions. Consistent with theoretical models, we find that strategy managers trade off these two higher moments when building optimal portfolios. Moreover, these trade-offs depend on the kind of strategy. Our experiments show that the VIX and its conditional variance are the most important factors affecting higher moment risk in the hedge fund industry. They also reveal that the behavior of hedge fund strategies is very asymmetric depending on the phase of the business cycle. In contrast to studies which rely on the mean-variance setting, we find that systemic risk—as measured by the cross-sectional dispersions of higher moments—tends to decrease in the low regime. The indicators of market volatility play a decisive role to explain this decline in systemic risk.

The subprime crisis was quite damaging for hedge funds. Using the local projection method (Jordà 2004, 2005, 2009), we forecast the dynamic responses of the betas of hedge fund strategies to macroeconomic and financial shocks—especially volatility and illiquidity shocks—over the subprime crisis in order to investigate their market timing activities. In a robustness check, using TVAR (Balke 2000), we simulate the reaction of hedge fund strategies’ betas in extreme scenarios allowing moderate and strong adverse shocks. Our results show that the behavior of hedge fund strategies regarding the monitoring of systematic risk is highly nonlinear in extreme scenarios—especially during the subprime crisis. We find that countercyclical strategies have an investment technology which differs from procyclical ones. During crises, the former seek to capture non-traditional risk premia by deliberately increasing their systematic risk while the later focus more on minimizing risk. Our results suggest that the hedge fund strategies’ betas respond more to illiquidity uncertainty than to illiquidity risk during crises. We find that illiquidity and VIX shocks are the major drivers of systemic risk in the hedge fund industry.

International audience; Predator-prey dynamics are widely used in ecology but seldom utilized in economics and marketing, despite their ability to express financial market agents' behaviors when considered in combination with economic cycles and financial crises. This multidisciplinary article presents a stylized framework of a market cycle that combines the notions of supply and demand and predator-prey interactions between buyers and sellers of housing mortgages. We illustrate our framework using data from the Global Financial Crisis and a Lotka-Volterra predator-prey model. We find that with our framework we can capture the dynamics of the market, particularly the peak and decline in the number of sellers and sold subprime mortgages. Our framework sheds a new light on consumer behaviors, pinpointing how they can put themselves into vulnerable prey positions. This article is one of the first of its kind to propose market phases and predator-prey dynamics nested in economic cycles and consumer buying trends.

We investigate the new Fama-French (FF, 2015, 2016) five factors augmented with a well-known illiquidity measure (Pastor and Stambaugh, 2003), using an innovative GMM robust instrumental variables estimator casted in a panel data framework. When using OLS, the augmented FF model seems to have explanatory power regarding the FF 12-sector returns. However, our panel data framework suggests that the only consistently significant factor is the market risk factor. Nevertheless, depending on the technique we use, we find that measurement errors may be the cause of this result, thus providing some empirical evidence in support of the new FF five-factor approach. As robustness checks, we also experiment with other liquidity measures – like the Amihud (2002) ratio and the term-spread – and bond-oriented factors. Across our 12 portfolios, the results are largely unchanged. We also apply our extended model to managed portfolios – i.e., hedge fund portfolios. The returns of hedge fund strategies seem more responsive to the augmented FF five-factor model that includes illiquidity measures, especially when accounting for the subprime crisis. There is also evidence that the new FF factors embed illiquidity.

We study the asymmetric responses of hedge fund return moments—especially higher moments as measured by return co-skewness and co-kurtosis—to macroeconomic and financial shocks depending on the phase of the business cycle. Similarly to previous papers on hedge fund systematic market risk (beta), we find that hedge funds seem to monitor their return co-skewness and co-kurtosis. The response of their return moments to VIX shocks—our indicator of macroeconomic and financial uncertainty—is particularly important, hedge funds reducing their beta and co-kurtosis and increasing their co-skewness following a (positive) VIX shock. Overall, the representative hedge fund tends to behave as an insurance seller in economic expansion and as an insurance buyer in recession or crisis. Finally, VIX shocks contribute to increase systemic risk in the hedge fund industry.

Increment variance reduction techniques are add-ons to Monte Carlo (MC) simulations. They make MC simulations converging faster by repeating the number of simulations with an incremental rate derived from mathematical functions. Besides speeding up MC simulations, the major advantage of increment techniques is their ability to handle large numbers of simulations avoiding memory saturation and overflow which occur when a plain MC simulation is involved in the pricing of multi-name credit derivatives. A trend among authors pricing financial securities with MC simulation has been to choose Quasi-Monte Carlo (QMC) methods using deterministic sequences instead of MC methods involving pseudorandom generators such as congruential generator and Mersenne twister. The Increment family models circumvent the constraint of identifying the optimal QMC sequence to price a given security by using a common generator such as Matlab-LCG-Xor RNG and determining the optimal mathematical function of incrementation of MC simulations that will make the pricing of the security adequate. Market participants in need of selecting a reliable numerical method for pricing complex financial securities such as multi-name credit derivatives will find our paper appealing.

We conjecture that market bubbles may be the results of the interplay of Goods and Bads (toxic products) which develop through three interlocking moments – herding, swarming and stampeding, with deviations marked by heteroscedasticity. We use our stylized model of financial predation, the Consolidated Model of Financial Predation, and data we have accumulated through in-the-field eight-year research and the study of 30 years of U.S. market history in order to explore the foundations of market crises. We find that blind trust (or the positivity bias) and of the fear to miss out on an opportunity to enter/exit a market impacts the investors’ decisions to invest or retract. We show how markets are driven towards a make-or-break predatory dynamic that creates winners and losers due in part to weak regulations and identify a constant k that permeates market behaviours.

Purpose The purpose of this paper is to test the new Fama and French (2015) five-factor model relying on a thorough sample of hedge fund strategies drawn from the Barclay’s Global hedge fund database. Design/methodology/approach The authors use a stepwise regression to identify the factors of the q-factor model which are relevant for the hedge fund strategy analysis. Doing so, the authors account for the Fung and Hsieh seven factors which prove very useful in the explanation of the hedge fund strategies. The authors introduce interaction terms to depict any interaction of the traditional Fama and French factors with the factors associated with the q-factor model. The authors also examine the dynamic dimensions of the risk-taking behavior of hedge funds using a BEKK procedure and the Kalman filter algorithm. Findings The results show that hedge funds seem to prefer stocks of firms with a high investment-to-assets ratio (low conservative minus aggressive (CMA)), on the one hand, and weak firms’ stocks (low robust minus weak (RMW)), on the other hand. This combination is not associated with the conventional properties of growth stocks – i.e., low high minus low (HML) stocks – which are related to firms which invest more (low CMA) and which are more profitable (high RMW). Finally, small minus big (SMB) interacts more with RMW while HML is more correlated with CMA. The conditional correlations between SMB and CMA, on the one hand, and HML and RMW, on the other hand, are less tight and may change sign over time. Originality/value To the best of the authors’ knowledge, the authors are the first to cast the new Fama and French five-factor model in a hedge fund setting which account for the Fung and Hsieh option-like trading strategies. This approach allows the authors to better understand hedge fund strategies because q-factors are useful to study the dynamic behavior of hedge funds.

The capital asset pricing model (CAPM), Fama-French (FF), and Pastor-Stambaugh (PS) factor models are examined using a new dynamic rolling regression version of the generalized method of moments (GMM) method. This rolling regression framework not only allows us to investigate phases of the business cycle, but also permits regression estimates to vary through time due to changes in the development and efficiency of the sectors. The principal reasons for using the dynamic GMM with robust instruments is that some of these factors are measured with errors and the disturbances may be non-spherical. The CAPM appears as the most parsimonious model to explain the FF sector returns. Furthermore, the rolling GMM approach is clearly more sensitive to dynamic financial episodes than the ordinary least squares approach. In particular, liquidity has some anticipatory power, as it is able to forecast the 2007–2009 crises with heightened volatility starting in late 2005.

We test the new Fama and French five-factor model on a sample of hedge fund strategies. This model embeds the q-factor asset pricing model which lies on the CMA and RMW factors. We find that the HML factor is not redundant for many strategies, as conjectured by Fama and French in their setting. HML seems to embed risk dimensions which are not included in the Fama and French new factors. In contrast to Fama and French, the α puzzle is robust to the addition of CMA and RMW. Furthermore, hedge funds seem to prefer, on the one hand, firms which invest a lot to firms which invest less, and, on the other hand, weak firms over robust ones. Finally, our results are not sensitive to the addition of the Fung and Hsieh seven-factor model. However, the explanatory power of the eleven-factor model is much higher for some hedge fund strategies involved in arbitrage.

We examine the behaviour of market agents during the years leading to the 2008 US subprime mortgage crisis using a stylized capital asset pricing model model. In our study, an average investor eager to make money by flipping houses meets a banker who offers him subprime mortgage deals. We refer to recent research that shows the mechanics of the psychological and behavioural components of these two market agents. In particular, much in line with the famous Stanford experiment, it is assumed that investors adopt a predator or a prey position. Our analysis shows that, given a historical tendency towards financial predatory acts on the part of market agents (including buyers), government regulations should be adapted and strengthened to face this dooming reality.

Nous examinons la performance du modele a cinq facteurs propose recemment par Fama et French (2015) dans le cadre des strategies suivies par les fonds de couverture. Du fait des aspects dynamiques de ces strategies, nous adoptons une procedure de changement de regime markovien dans laquelle les expositions aux facteurs de risque varient selon le regime – haussier ou baissier. Nous trouvons que l’ajout des facteurs qui commandent les rendements dans le modele q – a savoir le facteur relie a l’investissement des entreprises (CMA) et celui rattache a leur profitabilite (RMW) – n’ameliore pas de facon significative la performance globale du modele classique des rendements des fonds de couverture. Toutefois, nous trouvons que les facteurs CMA et RMW integrent des dimensions de risque qui ne sont pas captees par le facteur associe a la taille de l’entreprise (SMB) ou le facteur relie a la valeur d’une action (HML). Par ailleurs, quelques strategies reussissent a prevoir les chocs economiques et pilotent les facteurs de risque sur les deux regimes alors que d’autres eprouvent des difficultes a controler le risque au cours du regime de crise. Somme toute, et conformement aux autres etudes empiriques, nous trouvons que les facteurs de risque sont generalement plus actifs en regime baissier.

We introduce a versatile and robust model that may help policymakers, bond portfolio managers and financial institutions to gain insight into the future shape of the yield curve. The Burg model forecasts a 20-day yield curve, which fits a pth-order autoregressive (AR) model to the input signal by minimizing (least squares) the forward and backward prediction errors while constraining the autoregressive parameters to satisfy the Levinson–Durbin recursion. Then, it uses an infinite impulse response prediction error filter. Results are striking when the Burg model is compared to the Diebold and Li model: the model not only significantly improves accuracy, but also its forecast yield curves stick to the shape of observed yield curves, whether normal, humped, flat or inverted. Copyright © 2016 John Wiley & Sons, Ltd.

We investigate how hedge funds’ strategies react, as a group, to macroeconomic risk and uncertainty. Adopting the methodology of Beaudry et al. (2001), we track the behavior of the cross-sectional dispersions of hedge fund strategies’ returns, market betas and alphas over the business cycle. The pattern of strategies’ betas supports Beaudry et al.’s conjecture: hedge funds reduce their risk-taking (betas) during times of macroeconomic uncertainty, which makes their strategies more homogeneous and thus contributes to increased systemic risk in the financial system. However, the cyclical behavior of the cross-sectional dispersions of strategies’ returns and strategies’ alphas is not in line with Beaudry et al.’s conjecture. These dispersions tend to increase during episodes of rising macroeconomic uncertainty, which suggests the prevalence of the Black’s (1976) leverage effect during financial turmoil and the fact that the exposure of hedge fund strategies to risk factors is quite different from each other. Finally, although remaining important, procyclicality seems to have declined through time in the hedge fund industry, which suggests that a learning process is at play.

International audience; We advance the idea that the predator-prey dynamics that take place among key market agents play an important role in explaining financial crises. As such, we posit that financial markets evolve through fault lines involving toxic behaviors (such as deceit), toxic products (such as predatory mortgages) and inefficient regulations. We provide data to show that the puzzle of the lack of congruence between the market behaviors and what some economic models predict at times of financial crises may be the result of predator-prey interplays, and of so-called “predatory cells”, which are under the influence of financial accelerators.

Fama and French (FF, 2015) propose a five-factor asset pricing model that captures size, value, profitability and investment patterns. The primary purpose here is to further investigate this new model using an improved GMM-based robust instrumental variables technique. A further purpose is to explore the relationship among the FF factors and the Pastor–Stambaugh (PS, 2003) liquidity factor. We conclude that except for the market factor, all of the factors including liquidity are not significant at even the 5% level using our GMM approach for almost all of the FF 12 sectors.

This article uses a parsimonious and robust instrumental variables technique to minimize the specification errors in the Pastor-Stambaugh (PS) empirical model. In particular, we use an improvement of Hansen’s generalized method of moments (GMM) that uses higher moments that are robust instruments. We also propose a robustness test for these instruments. Results with these instruments indicate that the liquidity measure used in the PS empirical model is improperly measured and/or is ill-conceived. Although this article applies a GMM framework to a financial application, this technique is applicable to estimation problems in the presence of specification errors in all areas of quantitative finance.

A key issue in modelling conditional densities of returns of financial assets is the time-variation of conditional volatility. The classic econometric approach models volatility of returns with the generalized autoregressive conditional heteroscedasticity (GARCH) models where the conditional mean and the conditional volatility depend only on historical prices. We propose a new family of distributions in which the conditional distribution depends on a latent continuous factor with a continuum of states. The distribution has an interpretation in terms of a mixture distribution with time-varying mixing probabilities. The distribution parameters have economic interpretations in terms of conditional volatilities and correlations of the returns with the hidden continuous state. We show empirically that this distribution outperforms its main competitor, the mixed normal conditional distribution, in terms of capturing the stylized facts known for stock returns, namely, volatility clustering, leverage effect, skew...

Most of traded double barrier options are monitored in discrete time, their pricing being more challenging than in continuous time. A few solutions are analytical with a correction for continuity; the remaining solutions are numerical with lattices, grids or Monte Carlo (MC) simulation. We apply an original variance reduction technique to the pricing of European double barrier options monitored in discrete time. This technique speeds up significantly the MC simulation. The computational algorithm repeats each experiment with an increasing number of trials at a logarithmic rate and calculates a weighted average of the options values. We test the accuracy of the Login (Logarithmic Increment) model with an analytical solution adjusted for discretization and a naive numerical model using Monte Carlo simulation. We show that the Login model accurately prices double barrier options. Market participants in need of selecting a reliable numerical method for pricing double barrier options monitored in discrete time will find our article appealing. Moreover, the idea behind the method is simple and can be applied to the pricing of plain-vanilla or more complex derivatives, easing and speeding the valuation step significantly.

Econometricians have long recognized the need to account in some way for measurement errors, specification errors and endogeneity to ensure that the ordinary least squares estimator is consistent. This article introduces a new generalized method of moments estimator that relies on robust instruments to estimate panel data regression models containing errors in variables. We show how this GMM approach can be generalized for the panel data framework using higher moments and cumulants as instruments. The new instruments, engineered for greater robustness, are proposed to tackle the pervasive problem of weak instruments.

Using an updated database that extends after the subprime crisis, we revisit the asymmetries of hedge fund behavior in recession compared with economic expansion. In this respect, we study the time-varying α' sa ndβ's associated with strategy returns using an inno- vative framework based on the Kalman filter and the multivariate GARCH. We find that hedge fund managers reduce drastically their risk exposure during financial crises while their behavior is much smoother in normal times. We also find that hedge funds continue to provide good prospects for investors in terms of risk-adjusted returns. Actually, the procyclicality of hedge fund strategies' returns seems to decrease through time. Moreover, the strategies' behavior in terms of α and β tends to become more heterogeneous in times of crisis. The strategy exposure to adverse shocks seems to recede even after accounting for the subprime crisis. Finally, many hedge fund strategies benefit from an increase in the volatility of stock market returns. Hedge fund strategies may thus constitute a way to offset the lower expected returns observed in the conventional financial markets and may contribute to portfolio diversification.

We present an original Probabilistic Monte Carlo (PMC) model for pricing European discrete barrier options and compound real options. On the basis of Monte Carlo (MC) simulation, for barrier options the PMC model computes the probability of not crossing the barrier for knock-out options and crossing the barrier for knock-in options. This probability is then multiplied by an average sample discounted payoff of a plain vanilla option that has the same inputs as the barrier option but barrier-free, and to which we have applied a filter. We test the consistency of our model with an analytical solution (Merton, 1973; Reiner and Rubinstein, 1991) adjusted for discretization by Broadie et al (1997) and a naive numerical model using MC simulation presented by Clewlow and Strickland (2000). Our study shows that the PMC model accurately prices barrier options. Moreover, the idea behind the method is simple and can be applied to the pricing of complex derivatives, easing the valuation step significantly: we illustrate the versatility of the PMC model in pricing sequential compound real options. Market participants needing to select a reliable, versatile and simple numerical method for pricing options with embedded features will find our article appealing.

We revisit the factors incorporated in asset pricing models following the recent developments in financial markets – i.e., the rise of shadow banking and the change in the transmission channel of monetary policy. We propose two versions of the Fung and Hsieh (2004) hedge fund return model, especially an augmented market model which accounts for the new dynamics of financial markets and the procyclicality of hedge fund returns. We run these models with an innovative Hausman procedure, tackling the measurement errors embedded in the models factor loadings. Our empirical method also allows for confronting the drawbacks of the instruments used to estimate hedge fund asset pricing models.

We investigate conditional specifications of the five-factor Fama-French (FF) model, augmented with traditional illiquidity measures. The motivation for this time-varying methodology is that the traditional static approach of the FF model may be misspecified, especially for the endogenous illiquidity measures. We focus on the time-varying nature of the Jensen performance measure α and the market systematic risk sensitivity β, as these parameters are essentially universal in asset pricing models. To tackle endogeneity and other specification errors, we rely on our robust instrumental variables (RIV) algorithm implemented via a GMM approach. In this dynamic or time-varying conditional context, we generally find that the most significant factor is the market one, but illiquidity may matter depending on which states or estimation methods we consider. In particular, sectors whose returns embed a market illiquidity premium are more exposed to a binding funding constraint in times of crisis, which leads to deleveraging and a resulting decrease in systematic risk.

La iliquidez es bien conocida en la literatura como un importante factor de riesgo a considerar en los modelos financieros de retorno. Sin embargo, no hay mucho consenso sobre la medida que se debe utilizar como proxy para la iliquidez. Nuestras aportaciones a la cuestión se centran principalmente en la medida de Pástor-Stambaugh en el contexto del modelo tri-factorial de Fama-French y, más recientemente, en su nuevo modelo de cinco factores. En este artículo, se discuten las aportaciones anteriormente aludidas desde la perspectiva de errores en las variables. En particular, se proponen nuevos instrumentos robustos que se han desarrollado y aplicado en distintas etapas de nuestra investigación. En este artículo, se llevan a cabo los test de robustez de estos nuevos instrumentos. En general, nuestros nuevos instrumentos, con estimación GMM, muestran que tanto en el marco de sección cruzada o de datos panel como en una regresión recursiva/rolling regression (en comparación con el marco del filtro de Kalman) el factor más significativo es el factor de mercado. Estos resultados podrían interpretarse en la línea de la preocupación Cochrane acerca de un "zoológico de factores".

Traditional financial institutions as banks follow procyclical risk strategies, that is, they increase their leverage in economic expansions and reduce it in contractions, which leads to a procyclical behaviour for their betas and other risk and financial performance measures. We study the cyclical aspects of hedge fund strategies, an issue quite overlooked in the literature. We rely on two procedures: conditional modelling and Kalman filtering of hedge funds’ alpha and beta. We find that hedge fund betas are usually procyclical. Our results also show that the alpha is often high at the beginning of a market upside cycle but, as the demand pressure increases, it progressively shrinks, which suggests that the alpha puzzle ought to be analysed in a dynamic setting.

This article discusses the notion of perceived predation (the idea that a vis-a-vis wants us harm, by surprise) and reveals some of the key findings following a longitudinal study conducted with three different groups in 2011. The study shows that untrained people invited to negotiate financial transactions with another party naturally tend to adopt a strong stance, one that can be perceived as predatory, whereas those with training in negotiation adopt somewhat of a more conciliatory attitude. The important implication of such findings is that portfolio managers who attempt to maximize their client’s wealth in an unrestricted manner may well, consciously or not, turn against the very interests of those clients by using techniques that minimize perceived predation.

In this article, we propose a new empirical version of the Fama and French (F&F) model based on a new version of the Hausman (1978) specification test. This empirical model incorporates correction factors for risk exposure. These factors take into account the problem of errors-in-variables. Our model also features another innovation. It uses higher moments of the four factors of the augmented F&F model as instruments to tackle the problem of errors-in-variables. This new method reflects the nonlinear character of risk. We highlight the link between our new version of the Hausman artificial regression and the standard two-stage least-squares method. Our results suggest that it is preferable to account for errors-in-variables and more generally specification errors when estimating the F&F model because it might cause serious biases of important parameters like alpha and beta.

The findings presented in this article improve the existing methods for estimating financial models of returns and especially for estimating the parameters that are relevant for portfolio managers, such as the Jensen alpha, a popular measure for stock selection, and beta, a well-known systemic risk measure. The authors focus on the presence of measurement errors in these models. For instance, the risk factors in the Fama and French models or the market model are biased by measurement, or specification, errors. These measurement errors are related to the use of proxies for measuring risk factors, such as the risk premium, and other risk factors, which are approximated by mimicking portfolios. To tackle these specification problems, the authors propose new Hausman-based estimators lying on cumulants optimal instruments. Using these newly generated strong instruments in a generalized method of moments (GMM) setting, the authors obtain new GMM estimators that they call GMM-C and its homologue GMM-hm. They also extend the methodology to the standard two-stage least squares (TSLS) framework using new optimally generated instruments. These procedures improve the existing method of moments for estimating, or calibrating, the parameters of a financial model and, more generally, for treating the problem of endogeneity often encountered in financial studies. Moreover, this study leads to a new indicator that signals the presence of specification errors in financial models. The authors apply a battery of tests and estimators to a sample of 22 HFR hedge fund indices observed monthly over the period 1990–2005. Their tests reveal that specification errors bias parameter estimation of financial models of returns and that the ranking of hedge funds is very sensitive to the choice of estimators.

Estimation of financial models of hedge fund returns often gives rise to abnormally high alphas. This phenomenon may be due to mis-specified models, but recently the introduction of volatility factors in hedge fund returns models (Kuenzi and Xu, 2007) has been viewed as a solution to solve the alpha puzzle. This paper shows that modelling the volatility of the innovation term of hedge fund return models might be another way to explain the alpha puzzle. The model proposed in this paper is a factor model that incorporates an 'alternative' factor, which is the return of a short put on the Standard & Poor's 500 whose volatility is the VIX. This paper takes an overall view to analyse the problem of specification errors in financial models. To account for specification errors, it proposes a new estimator based on the generalised method of moments (GMM) whose instruments are the higher moments of returns, the GMM-hm. Some transformations of the basic factor model, which are recommended in the financial literature to improve the estimation of the alpha, are considered the n-CAPM and the conditional model. Some GARCH and EGARCH specifications of our basic model of returns are also estimated. Our results show that modelling the volatility of the innovation is the best way to lower the alpha. The n-CAPM has also had some success in reducing the alpha. The alpha is not the only garbage bag of a returns model (Jaeger and Wagner, 2005), but the alpha and the innovation both constitute the garbage bag.

This article proposes new estimations of conditional versions of the Fama-French model in order to deal with the alpha puzzle that emerges from hedge fund studies. Its originality lies in resorting to a new form of GMM, the GMM-C, which uses the cumulants of the explanatory variables as instruments in doing the estimations of each hedge fund strategy in panel. The estimations reveal that the new estimator is robust and is preferable to the usual OLS estimation of conditional models, which does not account for specification errors as does the GMM-C. This article also shows that the success of a conditional model in solving the alpha puzzle is related to a judicious choice of the conditioning information.

This paper proposes new Hausman-based estimators lying on higher moments and cumulants. Our study gives way to a new indicator signalling the presence of specification errors in financial models. We apply our battery of tests to a sample of 21 HFR hedge funds strategies over the period 1990–2005. Our tests reveal that it is much more preferable to account for specification errors when estimating financial models of returns. Actually, the performance ranking of hedge funds strategies may change significantly when accounting for specification errors.

A new literature has been recently devoted to the modeling of ultra-high-frequency (UHF) data. Our first aim is to develop an empirical application of UHF-GARCH models to forecast future volatilities on irregularly spaced data. We also compare the out-sample performance of these generalized autoregressive conditional heteroskedastic (GARCH) models with the realized volatility method. We propose a procedure to account for the time deformation problem and show how to use these models for computing daily Value at Risk (VaR).

The authors consider the dynamic market strategies of hedgefunds by using the Kalman filter. There are many studies onthe behavior of conditional alphas and betas of hedge funds,but the dynamics of these coefficients is studied within thetraditional regression framework: The resulting conditionalalphas and betas are hence for a great part arbitrary becausethey do not result from a dynamic optimization process. Inthis article, the authors try to correct this problem in partby writing transition equations for the alpha and the betawhose explanatory variables are market financial variables.The alphas of hedge fund indices appear quite difficult tocontrol, a result in line with the market efficiency hypoth-esis. Besides, the betas are much more controlable, their reac-tion to market variables being significant.

There is yet an unsolved beta puzzle in the financial literature. While studying the returns of mutual funds, Ferson and Schadt1 observed that these funds had a tendency to increase their conditional beta when the market risk premium decreases. According to them, the behaviour of the typical mutual fund is perverse and that gives way to the beta puzzle. Ferson and Schadt tried to explain this puzzle by the negative correlation between changes in conditional betas and change in net new money flows. Yet, cash holdings are positively correlated with net new money flows. Thus, when cash holdings are high, beta is low, a normal relation. Here, there is an endo-geneity econometric problem that must be dealt with a robust instrumental variable (IV) estimator.

This paper revisits an instrumental variable technique to minimize the errors-in-variables problem in capital asset pricing models. Our results show that Dagenais and Dagenais [Dagenais, M.G., Dagenais, D.L., 1997. Higher moment estimators for linear regression models with errors in the variables. Journal of Econometrics 76, 193–221] estimator, based on higher moments, is well suited to correct for the bias induced by measurement errors in both the CAPM and the three-factor model of Fama and French [Fama, E.F., French, K.R., 1997. Industry costs of equity. Journal of Financial Economics 43, 153-193].

This article uses a new set of instruments based on higher statistical moments to discard the specification errors that might be present in the Fama and French model. It shows that the usual instruments perform quite poorly in comparison to higher moments. It estimates the Fama and French model on a sample of 22 HFR hedge funds indices and 111 HFR individual hedge funds over the period 1990-2005. To do so, it compares many instrumental variables methods as the two-stage least squares and the generalized method of moments. The results show that there are few problems of specification errors on the side of the indices but this level of aggregation hides the errors. Indeed, the estimations of the sample of 111 funds reveal specification errors for the loadings of the market premium and the factor SMB which seem understated. The message to retain is that the individual investor must consider specification errors when modelling the returns of hedge funds as he does not buy the indices but the stocks of individual hedge funds. This article shows an investor how to correct the traditional measures of performance, which are alphas and factor loadings, to guide him towards a better decision process.