Vous êtes ici

Jason Prole

Diplômé DBA - 2020

Titre de thèse

The price of risk: the capital asset pricing model, short volatility and cointegration

Superviseur(s)

Carole Bernard
Achieving the required returns for liability-driven investors is an ongoing challenge. This research investigates the dominant framework for investing, the Capital Asset Pricing Model (CAPM). A price of risk in financial markets is a central requirement of the CAPM; however, identification of the appropriate investment index that reflects full diversification, and delivers the price for its risk, is unresolved. The result is uncertainty to both the returns and risks generated when using the CAPM and the effectiveness of mean-variance optimization for asset allocation. This research begins with investigating whether short futures on the CBOE volatility index (VIX) acts as the price of risk in equity markets and continues by asking whether a linkage exists between the price of equity risk and the price of credit risk in the fixed income markets. Finally, the research explores asset allocation for liability-driven investors by using a cointegration optimal asset allocation methodology that switches the focus from the short-term nature of mean-variance to the long-term focus on price-level.The methodology follows the five-factor framework of Fama and French (2015) using regression analysis to compare the explanatory ability of short volatility futures in describing equity market returns. It then proceeds with a similar investigation of the explanatory ability of short volatility futures for credit spreads, which parallels the work of Bai, Bali, and Wen (2018). Finally, the research investigates the effectiveness of cointegration to define the asset allocation for a liability-driven investor by paralleling the methodology of Alexander and Dimitriu (2005b) and the subsequent work of Chiu and Wong (2011). The results provide insight into market efficiency, cross-market integration, and liability-driven investing. This research suggests that VIX futures as the market factor deliver increased portfolio efficiency and non-market factor coefficients that are ex-ante more intuitive economically. Further, VIX futures proxy for the price of credit risk in diversified fixed income portfolios, while linking the equity and credit markets. In addition, the unbundling of risk factors improves portfolio efficiency for fixed income investors. Factor investors in equity and fixed income markets will find these novel outcomes economically beneficial.Cointegration optimal asset allocation equaled the performance of naïve mean-variance optimization (MVO) while increasing the probability of achieving ex-ante return expectations. In practice, these findings suggest the irrelevance of inputs assumptions for MVO, an economic rationale for discounting the liability with the expected return for US Public Pensions, and that managing time-varying return and risk improves portfolio efficiency. Key Words:CAPM; Factor Investing; Fixed Income; VIX Futures; Asset Allocation