Comparing the predictive efficiency of Fama-French Three-Factor Model and Carhart Four-Factor Model: Financial market approach

This study aims to compare the efficiency of predicting return on portfolios of the Fama-French Three Factor Model (FF3) and the Carhart Four Factor Model (CF4) using the Machine Learning approach. Experimental study on Hanoi Stock Exchange in the monthly cycle of 1/2015 to 12/2022. The securities are grouped into 10 diversified portfolios by beta every 6 months. The study divided the data into 2 train-test sets at the ratio of 70:30. The results show that, although the Root Mean Square Error (RMSE) of the CF4 model is lower than that of FF3, the T-test results show no difference between the two models.