Research on Dynamic Correlation Based on Stochastic Time-varying Beta and Stochastic Volatility
This paper uses monthly returns from 15 industry sectors in the Chinese stock market from 1/1/2011 to 30/12/2016 to study the characteristics of correlation between individual industry indices and a market portfolio on the prospective of stochastic timevarying beta and stochastic volatility. This paper also studies the response of correlation between industry indices and market portfolio to the variation of systemic risk (idiosyncratic risk) and market shock (idiosyncratic shock). This paper is
... ock). This paper is mainly focused on the following issues: a) Existence of Leverage Effect. b) Gradually decreasing correlation coefficient in response to an increase in idiosyncratic volatility. (It shows some inflection points due to some factors like market shock.) c) The asymmetric response of the correlation coefficient to positive and negative shock. d) Current period idiosyncratic volatility tends to increase in the next period, further decreasing the next period correlation coefficient. The model specification includes: a) Stochastic volatility modelling for state space variables. b) Cross sectional regression of dynamic correlation coefficients on market shock (idiosyncratic shock) and market volatility.