Semester

Spring

Date of Graduation

2011

Document Type

Dissertation

Degree Type

PhD

College

Chambers College of Business and Economics

Department

Economics

Committee Chair

Victor Chow.

Abstract

This dissertation comprises three essays that tackle various aspects of linear asset pricing models. The first essay, recognizing the dependence problem between factors, in the context of linear multi-factor models, proposes an optimal procedure to find orthogonalized risk premia, which also facilitates the decomposition of the coefficient of determination. Importantly, the new risk premia may diverge significantly from the original ones. The decomposition of risk allows one to explicitly examine the impact of individual factors on the return variation of risky assets, which provides discriminative power for factor selection. The procedure is experimentally robust even for small samples. Empirically we find that even though on average, approximately eighty (sixty-five) percent of style (industry) portfolios' volatility is explained by the market and size factors, other factors such as value, momentum and contrarian still play an important role for certain portfolios. The components of systematic risk, while dynamic over time, generally exhibit negative correlation between market, on one side, and size, value, momentum and contrarian , on the other side.;In the second essay we apply Marginal Conditional Stochastic Dominance (MCSD) tests to returns on sentiment-beta sorted portfolios and sentiment-arbitrage portfolios, constructed using the Baker and Wurgler (2007) index of sentiment levels. The theory of MCSD demonstrates that, if one (mutually exclusive) subset of a core portfolio dominates another, conditional on the return distribution of the core portfolio, then the core portfolio is inefficient for all utility-maximizing risk-averse investors. Based on returns on the U.S. equity market, we show that both positively and negatively sentiment sensitive stocks are conditionally and stochastically dominated by sentiment insensitive stocks. Moreover, we find dominance among sentiment-arbitrage portfolios, constructed with positively sensitive vs. insensitive, insensitive vs. negatively sensitive, and positively vs. negatively sensitive stocks. Therefore, we conclude that the market portfolio is stochastically inefficient.;The third essay builds on the theory of the FX market, which shows that investing in a high yielding currency would generate zero average returns, due to its depreciation with respect to the home currency. Empirically, it has been shown that high yielding currencies actually appreciate. The Consumption CAPM applied in the FX market argues that a high yielding currency also exposes investors to more consumption risk, hence the positive excess returns. The literature on the FX market proposes different strategies that generate abnormal returns based on the "forward puzzle". This paper investigates whether the returns obtained through a specific strategy that combines mean reversion and momentum in the FX market, are a compensation for risk. We find that the risk factors corresponding to an extended version of the Consumption CAPM can explain 99% of the variation in currency excess returns. We compare this model with other asset pricing models and conclude that none of them can provide a similarly good explanation.

Share

COinS