Testing alternative versions of the Fama-French three-factor model in frontier markets: The case of Jordan

Abstract

The objective of this study is to examine alternative versions of the Fama-French three-factor model on the Amman Stock Exchange (ASE). We consider two versions of the model: the Fama and French (1993) traditional model and the Cremers et al., (2012) modified model. Specifically, the study uses data on all stocks listed and traded in the regular stock exchange and covers the period from December 2000 through June 2017. To explore the size and book-to-market effects, we form nine portfolios sorted on size/book-to-market. The size/book-to-market portfolios are constructed from an independent sort and rebalanced annually. To evaluate the models, the study applies the Ordinary Least Squares (OLS) time-series regressions approach of Black et al., (1972). The study finds that size and book-to-market effects exist in the ASE equity market returns. The asset pricing tests show that small-size and high-growth portfolios, respectively, appear riskier than big-size and low-growth portfolios and that neither the traditional model nor the modified model is successful in capturing the size and book-to-market effects, however, the traditional model better describes the returns than the modified model on the ASE. The important implications of the results are that for academic researchers, portfolio managers, and individual investors, any applications of the models on the ASE should be made with caution, and that size and value factors construction methods do matter when evaluating the model on the ASE.