Econometric Analysis of Continuous Time Asset Pricing Models
报告人: Yoosoon Chang, Indiana University
时间:2015-12-09 14:00 ~ 15:30
地点:Room 217, Guanghua Building 2
Abstract:
We develop an econometric theory and methodology for general continuous time asset pricing regression models derived under no arbitrage condition. To allow for general stochastic volatilities in these regressions, considered are two estimation approaches, one based on the random sampling scheme using a volatility time and the other relying on the direct correction for heteroskedasticity by generalized least squares. Both of them are asymptotically equivalent to the continuous time maximum likelihood estimator. However, they have nonstandard limit distributions, and render the conventional tests to be invalid, if the regressors are persistent and not strictly exogenous. To deal with this problem, we introduce an additional procedure for endogeneity correction, which provides fully efficient inference with standard Gaussian limit theory. Some empirical illustrations as well as simulation results are also presented.
About the Speaker: