Essays on international asset pricing anomalies

It is well known that the traditional asset pricing theory fails to explain the existence of value anomaly for both the domestic and international markets. In this thesis I examine the international value anomaly from different angles. The first chapter tests the validity of the traditional CAPM with a two-beta dynamic conditional volatility model which has some success explaining the value anomaly with domestic data. I conclude that even a conditional volatility model still cannot explain the value anomaly and the traditional CAPM does not hold well either. The second chapter further examines the same question from a distributional perspective. In this chapter I use stylized copula models to capture the dependence structure between the test portfolio and the market factor in order to explore the deficiency of the traditional normality assumption used by the CAPM. I find that the normal distribution assumption does not fit the data well in both unconditional and conditional cases. The third chapter investigates the comovement between different country portfolios instead of a test portfolio and the market factor because it is very important to understand how financial series move together in order to capture the potential risk sources in the market. This chapter concludes the existence of the downside risk between countries and more so between the developed countries than the emerging markets.