Quantitative modeling of financial markets: four tests of the three-factor model in the French case.

Summary The limitations of empirical applications of the CAPM have raised the issue of anomalies. The objective of the research conducted is to better understand equity returns by distinguishing risk factors from anomalies. The thesis focuses on the three-factor model of Fama and French (1993) and its application to the French market. Over the period from July 1976 to June 2001, the addition of the HML and SMB factors to the market portfolio provides a better explanation of French stock returns, both in time series and in cross-section. The explanatory power of the three-factor model (31.45%) is on average three times higher than that of the CAPM (9.72%). The study of the Daniel and Titman (1997) proposal, which refutes the hypothesis of risk factors, leads to contrasting results in the French case. Some methodological difficulties are mentioned. Another alternative hypothesis that has been tested is to circumvent the limitations of the CAPM by adding the third- and fourth-order co-moments to the three-factor model. An additional empirical investigation testing the market portfolio misspecification hypothesis of Ferguson and Shockley (2003) is conducted. The main conclusion of this study is that the true market portfolio is not a substitute for the SMB and HML portfolios in explaining returns in both time series and cross-section. Finally, a last result relating to the non-dependence of the explanatory capacity of the three-factor model on economic conditions is put forward. The three-factor model seems to incorporate information on economic activity.