Estimating the Asymmetric Pattern of Volatility in China Stock Markets. A New Proposal

摘要

Over the past 25 years, volatility models and their forecasts have been the focus of both academic researchers and practitioners. However, the models proposed in the literature use to fail to detect the asymmetric pattern of volatility in a large number of cases. This is the reason why in this paper we propose a new asymmetric ARSV model: the Threshold Asymmetric Autoregressive Stochastic Volatility (TA-ARSV) model. We focus our empirical study on China's daily Stock Exchange returns. The are four main reasons for this: China can be considered a relatively "new" market; the growth potential of Chinese stock markets in emerging countries has attracted qualified foreign institutional investors and global investors in recent years; Chinese stock markets have experienced rapid growth associated with high risk; and there has been relatively little work done on modelling and forecasting stock market volatility in China with inconclusive results.