Factor Models in China: How Well Do Factor Models Describe Equity Returns? An Empirical Study on Factor Models in the Chinese Equity Market

Henrik Aalberg & Lukas Schubart

Student thesis: Master thesis


This master thesis concerns itself with factor effects in the Chinese equity market. In order to answer how well factor models explain equity returns in the Chinese market, this thesis tests the CAPM of Sharpe (1964) and Linter (1965), the three-factor model of Fama and French (1993), the four-factor model of Carhart (1997), and the five-factor model of Fama and French (2015) in the Chinese A-share market. In addition to these models, the individual factors that are a part of the four different models are also assessed. The assessment of models and factor effects are performed by means of descriptive methods, as well as univariate regression, multivariate regression, and correlation methods. Based on the above-described methods, the models are tested in line with the framework developed by Fama and French (1993, 2015). A comparison of standard measures of fit and the more comprehensive GRS tests are performed across the models. The resulting findings are that the four- and five-factor models best fit the A-share market. The factors and their overall interplay seem to explain much of the variation in the Chinese equity market. For the multifactor models tested, three out of four sorts performed are unable to reject the null hypothesis of the GRS test, and the average R-squared for each model is approximately 0.92. For the individual factor effects, it can be seen that the market premium is positive but insignificant, the Size premium is strong and significant, and there seems to be a lack of a B/M, profitability, and investment premia. Furthermore, a significant medium-term reversal effect is present in the market rather than a momentum effect. These findings contrast comparable research on global markets but are broadly in line with research on the Chinese market.

EducationsMSc in Applied Economics and Finance, (Graduate Programme) Final Thesis
Publication date2018
Number of pages120