Explaining anomalies in Australia with a five-factor asset pricing model

This paper compares the ability of three-factor and five-factor asset pricing models to explain the apparent profitability of a broad selection of anomalies in Australian equity returns. Rather than examining the fit of each model to common test portfolios, our focus is on the spread return to long-short trading strategies designed around so-called anomalies. After documenting significant spread returns to 16 anomalies (including several not previously studied in Australia), the empirical analysis provides cautious support that the recently-proposed investment and profitability factors have a role to play. The number of anomalies that remains after risk adjustment decreases under the five-factor model. Further, while the magnitude of reduction in alpha is modest, our testing shows that it is statistically significant in many cases. However, both three-factor and five-factor models repeatedly fail the Gibbons, Ross, and Shanken's (1989) (GRS) test, suggesting that the quest for a better asset pricing model is not yet complete.