I use a regime-switching generalization of principal components analysis (PCA) to estimate costs of equity for a set of 48 industry portfolios. In many cases, the cost-of-equity estimates display substantial variation across the unobserved regimes. Furthermore, the estimated pricing errors produced by the PCA factors are smaller in magnitude than those for factor pricing models that feature prominently in the asset-pricing literature. In regression-based tests, for example, the factor-mimicking portfolios for a six-factor PCA specification produce a mean absolute estimated intercept (alpha) that is 45% smaller than the mean absolute estimated intercept for the Fama and French (2015) five-factor model.
I propose a regime-switching generalization of instrumented principal components analysis (IPCA) that yields new insights about the relation between characteristics, factor loadings, and expected stock returns. Using a two-regime specification, I find evidence of a high-volatility regime in which individual stocks have high conditional expected returns. This contrasts sharply with the pattern of bull and bear regimes that is obtained by analyzing only market returns. Although exact factor pricing can be rejected, characteristics are more strongly related to priced covariances in the high-volatility regime. Furthermore, regime-switching predictability makes a substantial incremental contribution to the out-of-sample explanatory power of IPCA estimates.
We document a pronounced negative interaction between short-term return reversals and prior trading activity for both small-cap and large-cap stocks. Stocks with low prior turnover display a strong monthly reversal effect. In contrast, those with high prior turnover display a monthly continuation effect (short-term momentum). Motivated by existing models of the trading process, we posit that the observed interaction stems from the relation between turnover and news that spurs speculative trading. We investigate this proposition empirically using a simple proxy for the fraction of turnover that is driven by news. The results are consistent with the predictions of our hypothesis.
I analyze the cross-section of covariance risk for individual stocks using a new type of multivariate volatility model in which firm characteristics serve as time-varying loadings on fundamental factors. The evidence points to strong linkages between firm characteristics and covariance risk, and also reveals that cross-sectional differences in covariance risk explain much of the cross-sectional variation in expected excess stock returns. I find, for example, that the fundamental factors perform at least as well as the Fama-French factors in regression-based pricing tests. In view of its tractability and performance, the proposed model should find use in a variety of applications.
We propose a comprehensive empirical strategy for optimizing the out-of-sample performance of sample mean-variance efficient portfolios. After constructing a sample objective function that accounts for the impact of estimation risk, specification errors, and transaction costs on portfolio performance, we maximize the function with respect to a set of tuning parameters to obtain plug-in estimates of the optimal portfolio weights. The methodology offers considerable flexibility in specifying objectives, constraints, and modeling techniques. Moreover, the resulting portfolios have well-behaved weights, reasonable turnover, and substantially higher Sharpe ratios and certainty-equivalent returns than benchmarks such as the 1/N portfolio and S&P 500 index
Firm Characteristics, Cross-Sectional Regression Estimates, and Asset Pricing Tests, forthcoming in Review of Asset Pricing Studies
The Value Premium and Expected Business Conditions, Finance Research Letters 30, 2019
Estimating the Cost of Equity Capital Using Empirical Asset Pricing Models, International Review of Finance 19, 2019
Income Shifting as an Aspect of Tax Avoidance: Evidence from U.S. Multinational Corporations, with A. Cordis, Review of Pacific Basin Financial Markets and Policies 21, 2018
Capital Expenditures and Firm Performance: Evidence from a Cross-Sectional Analysis of Stock Returns, with A. Cordis, Accounting and Finance 57, December 2017
Discrete stochastic autoregressive volatility, with A. Cordis, Journal of Banking and Finance 43, June 2014
Component-driven regime-switching volatility, with J. Fleming, Journal of Financial Econometrics 11, Spring 2013
It’s all in the timing: Simple active portfolio strategies that outperform naive diversification, with B. Ostdiek, Journal of Financial and Quantitative Analysis 47, April 2012
Regime-switching factor models in which the number of factors defines the regime, with A. Cordis, Economics Letters 112, August 2011
Long memory in volatility and trading volume, with J. Fleming, Journal of Banking and Finance 35, July 2011
Most Frequently Cited Articles
The economic value of volatility timing using ‘realized’ volatility, with J. Fleming and B. Ostdiek, Journal of Financial Economics 67, March 2003
The economic value of volatility timing, with J. Fleming and B. Ostdiek, Journal of Finance 56, February 2001
Information and volatility linkages in the stock, bond, and money markets, with J. Fleming and B. Ostdiek, Journal of Financial Economics 49, July 1998