Dynamic portfolio choice: Time-varying and jumps

Abstract

Assuming that the risky asset return follows stochastic j ump-diffusion model, this paper studies the effect of the time-varying and jumps of return process on dynamic portfolio choice, and discusses their characteristics in Chinese stock market. It applies stochastic control theory to obtain the analytical solution of dynamic portfolio choice, which maximizes the expected power utility of portfolio terminal wealth, and utilizes the Spectral Generalized Method of Moments (SGMM) to estimate model parameters to do an empirical study based on the monthly data of Shanghai Exchange Composite Index from 1997.01–2008.12. Results show, the time-varying results in the positive or negative intertemporal hedging demand, which depends on the degree of investor's risk-aversion and the correlation coefficient between the shift of risky asset return and that of risk premium; the jumps also results in positive or negative jump hedging demand, but overall reduces the position on risky asset; as for the stock market in China, the effect of jumps is greater than that of the time-varying, and their horizon effect is very weak.