Building a cross-sectional asset pricing model
The ASSETPRICINGPUZZLES (Cross sectional asset pricing puzzles: An equilibrium perspective) project focused on developing an inter-temporal asset pricing model that uses a cross-section of firms that are characterised by mean-reverting expected dividend growth. Substantial empirical data was found supporting the cross-sectional implications of the model and market anomalies were generally found to not be contradictory to rational asset pricing. Following with this research, the commonalities across asset pricing anomalies were explored. Evidence proved the profitability of strategies, such as price and earnings momentum, occurs when short positions are taken in high-credit-risk firms that have worsening credit conditions. The entirety of this part of the project was published in the Journal of Financial Economics and won the Fama-DFA Prize for best paper. International asset pricing was also studied, in addition to studying domestic (United States) asset pricing models. This aspect of the project proposed a risk-based explanation of certain deviations from global asset pricing that had been unresolved until now by other global asset pricing models. Furthermore, the inter-temporal asset pricing model was used for asset allocation purposes. Interestingly, the common reduced-form approach to analysing asset allocation was compared and contrasted using the vector autoregression through which investors forecast asset returns. Ultimately, it was discovered that the long-risk investor's gains come from their ability to avoid exposure to large negative events and the reduced-form investor takes more advantage of periods of high average returns. Additionally, the research was used to create a simple dividend discount model. Through this model, contrasting evidence of a large negative versus a large positive relation between idiosyncratic volatility and average returns can be resolved. Finally, three working papers are currently being prepared to use the asset pricing anomaly-based literature to undertake three studies. Using a measurement at the stock level of stock overpricing, these studies aim to predict mutual fund and hedge fund performance, and detect mispricing in call and put options. These findings and their continued development and study will aid understandings of cross-sectional asset pricing and improve the related ability to predict asset returns.
Keywords
Cross-sectional asset pricing, asset pricing model, asset returns, fund performance