Forecasting the Ability of Dynamic versus static CAPM: Evidence from Amman Stock Exchange

This study tests whether the dynamic (conditional) Capital Asset Pricing Model (CAPM) outperforms the staticone in forecasting the returns of the industrial companies listed in Amman Stock Exchange (ASE) over theperiod 2000-2011. We investigate the in-sample forecasting ability of CAPM estimated via OLS, GJR-GARCH(1, 1), and Kalman Filter. The results indicate that the dynamic CAPM estimated through GJR-GARCH (1, 1)provide the most accurate in-sample forecasts of stock returns. Moreover, this model shows the lowest values ofAkaike Information Criterion and explains the cross section of returns of most sample stocks