Estimation and application of the multifactor asset pricing models



Journal Title

Journal ISSN

Volume Title



In this dissertation, two articles examine the Carhart's four-factor (1997), Fama French five-factor (2015), and Fama French six-factor (2017) asset pricing models for six different portfolios in six different regions. In first article, the Fama-French six-factor model is tested for its ability to explain and forecast investment returns in six distinct portfolios across six regional stock markets. The empirical study yields key general conclusions. First, markets are efficient in that they do not allow for anomalous excess returns. (2) The six-factor Fama-French risk model has no redundant risk variables. Third, there are unknown idiosyncrasies in Japan, Asia, and developing markets. Finally, the ARIMAX model outperforms the ARIMA model in forecasting excess returns. It is shown in second article how the asset pricing models of Carhart's four-factor, Fama-French five-factor and Fama-French six factor affect the excess returns of various portfolios in various markets. With monthly data from Fama-French from January 2000 to October 2020, the multifactor asset pricing models are estimated using pooled OLS and three specifications of panel data fixed effects models. This data was gathered for six different portfolios in six different regions. The six portfolios are created by combining two different market capitalization sizes (small and big) with three different book-to-market equity ratios (low, medium, and high). The stock markets of the US, North America, Europe, Japan, Asia Pacific, and emerging regions, are used to assess the performance of each of the six portfolios in turn. A number of significant findings are obtained via empirical investigation. One of the most important points to note is that none of the factors in the Fama-French six-factor (and other) asset pricing models are redundant when it comes vi to explaining returns for different portfolios in different markets, despite the fact that their effects may differ across different portfolios. Second, anomalous returns (either positive or negative) vary among portfolios and geographies, presenting chances for investors to profit from these variations. Third, the COVID-19 pandemic had an impact on the results of certain portfolios, but not all of them were impacted. Lastly, for various portfolios, different specifications of the empirical models are preferred to one another. Lastly, when using panel data fixed effects requirements for multifactor asset pricing models, the explanatory power of the models is greater than when using the specifications utilized in prior research

Restricted until 06/2027. To request the author grant access, click on the PDF link to the left.



Asset Pricing Models