Estimation of Risk Premia with Omitted Variable Bias : Evidence from China

Standard estimates of risk premiums in traditional linear asset pricing models are subject to bias, primarily because of the omission of certain factors. The three-pass method is a promising approach that has been proposed to estimate the risk premium of observable factors. This method remains valid even if all relevant factors are not explicitly specified or observed in the model. We apply this method to construct a portfolio using Chinese A-share market stocks as the test asset. Empirical findings reveal that the three-pass method may be more effective in estimating risk premiums than traditional linear asset pricing models