Simulation-based estimation methods for cross-sectional financial asset pricing

This paper extends the simulation-based estimation method proposed by Phillips and Yu (2009) to the cross-sectional case. We examine their finite-sample performance by conducting Monte-Carlo simulations of this simulation-based method to both the time-series model and the cross-sectional model. The simulation results show that the proposed simulation based estimator can always reduce the percentage bias over the respective MLE and OLS estimator. Meanwhile, they do not significantly increase the variance or RMSE over their correspondent MLE and OLS estimator.