Tests of conditional asset pricing models in the Brazilian stock market

In this paper, we test a version of the conditional CAPM with respect to a local market portfolio, proxied by the Brazilian stock index during the period 1976–1992. We also test a conditional APT model by using the difference between the 30-day rate (Cdb) and the overnight rate as a second factor in addition to the market portfolio in order to capture the large inflation risk present during this period. The conditional CAPM and APT models are estimated by the Generalized Method of Moments (GMM) and tested on a set of size portfolios created from individual securities exchanged on the Brazilian markets. The inclusion of this second factor proves to be important for the appropriate pricing of the portfolios.     

Government Bond Returns,Measurement of Interest Rate Risk,and the Arbitrage Pricing Theory

Empirical tests are reported for Ross' arbitrage pricing theory using monthly data for U.S. Treasury securities during the 1960–1979 period. We find that mean returns on bond portfolios are linearly related to at least two factor loadings. Multivariate test results, however, are not consistent with the APT. Our sample data in the U.S. Treasury securities market are also not consistent with either version of the CAPM. One-month-ahead forecasts of excess returns using factor-generating models are compared with corresponding naive predictions or predictions using the “market model” with various market portfolios.     

A conditional asset-pricing model with the optimal orthogonal portfolio

This paper employs a conditional asset-pricing model based on the optimal orthogonal portfolio approach to construct a factor portfolio that embodies all the latent factors important for pricing a given set of test assets. The advantage of this portfolio to the anomaly related mimicking portfolios is its ability to separate out the components of average return that are not related to the return covariation. The performance of this portfolio is evaluated against several conventional factors, using both cross-sectional and time-series regression approaches, as well as the Hansen and Jagannathan (1997) distance measure. Its strong out-of-sample results indicate that our suggested methodology may have important applications in risk management, portfolio selection and performance evaluation.     

Changes in Expected Security Returns,Risk, and the Level of Interest Rates

Regressions of security returns on treasury bill rates provide insight about the behavior of risk in rational asset pricing models. The information in one-month bill rates implies time variation in the conditional covariances of portfolios of stocks and fixed-income securities with benchmark pricing variables, over extended samples and within five-year subperiods. There is evidence of changes in conditional “betas” associated with interest rates. Consumption and stock market data are examined as proxies for marginal utility, in a general framework for asset pricing with time-varying conditional covariances.     

Information Uncertainty and Post-Earnings-Announcement-Drift

Abstract:  We examine whether rational investor responses to information uncertainty (IU) explain properties of and returns to the post-earnings-announcement-drift (PEAD) trading anomaly. Consistent with a rational learning explanation, we find that: (1) unexpected earnings (UE) signals that are characterized as having greater IU have more muted initial market reactions; (2) extreme UE portfolios are characterized by securities with higher IU than non-extreme UE portfolios; and (3) within the extreme UE portfolios, high IU securities are more prevalent and earn larger abnormal returns than low IU securities. Further tests show that prior evidence of greater PEAD profitability for higher idiosyncratic volatility securities is explained by the greater information uncertainty associated with these securities.     

Testing conditional factor models

Using nonparametric techniques, we develop a methodology for estimating and testing conditional alphas and betas and long-run alphas and betas, which are the averages of conditional alphas and betas, respectively, across time. The estimators and tests can be implemented for a single asset or jointly across portfolios. The traditional Gibbons, Ross, and Shanken (1989) test arises as a special case of no time variation in the alphas and factor loadings and homoskedasticity. As applications of the methodology, we estimate conditional CAPM and multifactor models on book-to-market and momentum decile portfolios. We reject the null that long-run alphas are equal to zero even though there is substantial variation in the conditional factor loadings of these portfolios.     

CONDITIONAL HETEROSCEDASTICITY IN THE MARKET MODEL AND EFFICIENT ESTIMATES OF BETAS

Previous studies have investigated only unconditional heteroscedasticity in the market model. This paper tests for both conditional and unconditional heteroscedasticities as well as normality. Using the monthly stock rate of return data secured from the Center for Research in Security Prices (CRSP) tape for 1976 through 1983, this paper shows that conditional heteroscedasticity is more widespread than unconditional heteroscedasticity, suggesting the necessity of model refinements that take conditional heteroscedasticity into account. This paper provides an alternative estimation of betas of individual securities and portfolios based on the autoregressive conditional heteroscedastic (ARCH) model introduced by Engle. The efficiency of the market model coefficients is markedly improved across all firms in the sample through the ARCH technique.     

Exploring the Conditional Performance of U.K. Unit Trusts

We evaluate the conditional performance of U.K. equity unit trusts using the approach of Lynch and Wachter (2007, 2008) relative to three conditional linear factor models. We find significant time variation in the conditional performance of some trust portfolios and individual trusts using the lag term spread as the information variable. The conditional performance of the trusts is countercyclical and larger trusts have more countercyclical performance than smaller trusts within certain investment sectors. These patterns in conditional trust performance cannot be fully explained by the underlying securities that the trusts hold.

Asset pricing models and economic risk premia: A decomposition

The risk premia of linear factor models on economic (non-traded) risk factors can be decomposed into: i) the premium on maximum-correlation portfolios mimicking the factors; ii) (minus) the covariance between the non-traded components of the pricing kernel and the factors; and iii) (minus) the mispricing of the maximum-correlation portfolios. For a given set of assets available for investment, the first component is the same across models and is typically estimated with little bias and high precision. We conclude that the premia on maximum-correlation portfolios are appealing alternatives to the risk premia of linear factor models, with the dividend yield being the only economic factor significantly priced.     

Benchmark Invariancy,Seasonality and APM-Free Portfolio Performance Measures

The Jobson-Korkie (1981) Z score and the positive period weighting (PPW) score of Grinblatt and Titman (1989) are applied to various benchmarks of market and mimicking portfolios to study the benchmark invariancy problem. Significantly different portfolio performance inferences are found for a sample of 146 equity mutual funds depending on the mean-variance efficiency of the portfolio benchmarks (mimicking portfolios versus market indices). Portfolio performance inferences are affected significantly by the number of factors, nonsynchronous trading adjustment, and the sizes of the firms used for factor extraction. The returns of the portfolio benchmarks exhibit significant monthly seasonalities, which, in turn, significantly influence mutual fund performance inferences.     

Cash, investments and asset returns

We use an investment-based asset pricing model to examine the effect of firms’ investments relative to cash holdings on stock returns, assuming holding cash lowers transaction costs. We find that mimicking portfolios based on investments relative to non-cash capital and based on investments relative to cash capital are priced for various testing portfolios. On average, momentum stocks and growth stocks are more sensitive to the factor constructed using investment relative to cash.     

OPTIMAL PORTFOLIOS: MARKOWITZ FULL COVARIANCE VERSUS SIMPLE SELECTION RULES

Two major problems faced by portfolio managers are estimating the risk and return characteristics of individual securities and combining individual security risk and return estimates into optimal portfolios. The second problem is investigated in this paper by using the simple ranking criteria suggested by Elton, Gruber, and Padberg (EGP). The empirical results indicate that the EGP procedure is effective in estimating Markowitz efficient portfolios and can be an effective screening procedure for large numbers of securities.     

Correlation and Volatility Dynamics in International Real Estate Securities Markets

We study international correlation and volatility dynamics of publicly traded real estate securities using monthly returns from 1984 and 2006. We also examine, for comparison, the correlations among the corresponding stock markets. A multivariate dynamic conditional correlation model captures the time-varying correlation within the full period. We confirm lower correlations between all real estate securities market returns than those between the stock market returns themselves. Some significant variations and structural changes in the correlation structure happened within the sample period. We detect a strong and positive connection between real estate securities market correlations and their conditional volatilities. We also find the international correlation structure of real estate securities and the broader stock market are linked to each other. Our results have economic motivations regarding the potential integration of international real estate securities markets and the possibility of including information on changing correlations and volatilities to design more optimal portfolios for international real estate securities.

Testing conditional factor models: A nonparametric approach

Some recent studies of conditional factor models do not specify conditioning information but use data from small windows to estimate the time series of conditional alphas and betas. In this paper, we propose a nonparametric method using an optimal window to estimate time-varying coefficients. In addition, we offer two empirical tests of a conditional factor model. Using our new method, we examine the performance of the conditional CAPM and the conditional Fama-French three-factor model in explaining the return variations of portfolios sorted by size, book-to-market ratios, and past returns, for which recent literature has generated controversial results. We find that, although in general the conditional FF model outperforms the conditional CAPM, both models fail to explain well-known asset-pricing anomalies. Moreover, for both models, the failure is more pronounced for the equally-weighted portfolios than for the value-weighted ones.     

THE CROSS-AUTOCORRELATION OF SIZE-BASED PORTFOLIO RETURNS IS NOT AN ARTIFACT OF PORTFOLIO AUTOCORRELATION

Prior studies find evidence of asymmetric size-based portfolio return cross-autocorrelations where lagged large firm returns lead current small firm returns. However, some studies question whether this economic relation is independent of the effect of portfolio return autocorrelation. We formally test for this independence using size-based portfolios of New York Stock Exchange and American Stock Exchange securities and, separately, portfolios of Nasdaq securities. Results from causality regressions indicate that, across all markets, lagged large firm returns predict current small firm returns, even after controlling for autocorrelation in small firm returns. These cross-autocorrelation patterns are stronger for Nasdaq securities.     

Size and book-to-market factors in a multivariate GARCH-in-mean asset pricing application

The analysis of Fama and French (1993), is broadened to include time varying risk. This is achieved by an MGARCH-m application which extends the approach taken by Bollerslev et al. (1988) to a multiple index context. Appropriate weightings of the conditional cross-moments of returns on portfolios that make-up factor proxies are modelled as MGARCH. The mean equation of the model is designed to determine whether proxied sources of time-varying non-diversifiable risk can explain movements in excess returns on various types of portfolios. Time-variation in conditional excess return appears only to have a significant relation with time-varying conditional variance associated with a book-to-market factor.     

Approximate Factor Structures: Interpretations and Implications for Empirical Tests

This paper provides some new insights about approximate factor structures, as defined by Chamberlain and Rothschild [2], and their implications for empirical tests. First, we show that any economy that satisfies an approximate factor structure can be transformed, in a manner that does not alter the characteristics of investor portfolios, into an economy that satisfies an exact factor structure, as defined by Ross [9]. Second, we show that principal components analysis represents just one of many methods of forming groups of well-diversified portfolios with no idiosyncratic risk in large samples. Correct factor loadings will be obtained by regressing security returns on any group of these portfolios. Our interpretations of the Chamberlain and Rothschild results also provide additional insights into the testability of the Arbitrage Pricing Theory. We show that securities cannot be repackaged to hide factors in the manner suggested by Shanken [10] without the variance of some of the repackaged securities approaching infinity in large economies.     

CAPM over the long run: 1926–2001

A conditional one-factor model can account for the spread in the average returns of portfolios sorted by book-to-market ratios over the long run from 1926 to 2001. In contrast, earlier studies document strong evidence of a book-to-market effect using OLS regressions over post-1963 data. However, the betas of portfolios sorted by book-to-market ratios vary over time and in the presence of time-varying factor loadings, OLS inference produces inconsistent estimates of conditional alphas and betas. We show that under a conditional CAPM with time-varying betas, predictable market risk premia, and stochastic systematic volatility, there is little evidence that the conditional alpha for a book-to-market trading strategy is different from zero.     

Creating Fama and French Factors with Style

This paper utilizes Frank Russell style portfolios to create useful proxies for the Fama and French (1992) factors. The proxy‐mimicking portfolios are shown to represent a pervasive source of exposure across U.S. industry portfolios and to generally possess similar properties to those utilized in the finance literature. Further, a set of multivariate asset‐pricing tests of the three‐factor Fama and French asset‐pricing (FF) model based on the proxy factors fails to reject the model. However, these tests do not reveal strong evidence of significantly positive risk premiums, particularly in the case of the size and book‐to‐market factors.     

Liquidity risk,leverage and long-run IPO returns

We examine the risk-return characteristics of a rolling portfolio investment strategy where more than 6000 Nasdaq initial public offering (IPO) stocks are bought and held for up to 5 years. The average long-run portfolio return is low, but IPO stocks appear as “longshots”, as 5-year buy-and-hold returns of 1000% or more are somewhat more frequent than for non-issuing Nasdaq firms matched on size and book-to-market ratio. The typical IPO firm is of average Nasdaq market capitalization but has relatively low book-to-market ratio. We also show that IPO firms exhibit relatively high stock turnover and low leverage, which may lower systematic risk exposures. To examine this possibility, we launch an easily constructed “low-minus-high” (LMH) stock turnover portfolio as a liquidity risk factor. The LMH factor produces significant betas for broad-based stock portfolios, as well as for our IPO portfolio and a comparison portfolio of seasoned equity offerings. The factor-model estimation also includes standard characteristic-based risk factors, and we explore mimicking portfolios for leverage-related macroeconomic risks. Because they track macroeconomic aggregates, these mimicking portfolios are relatively immune to market sentiment effects. Overall, we cannot reject the hypothesis that the realized return on the IPO portfolio is commensurable with the portfolio's risk exposures, as defined here.