Effective Transfer Entropy Approach to Information Flow Among EPU, Investor Sentiment and Stock Market

Abstract

Although transfer entropy can test the nonlinear causal relationship between sequences, it is mainly used for stationary data. For nonstationary sequences with large fluctuations, the traditional transfer entropy method has obvious defects. Based on traditional transfer entropy, this paper proposes a transfer entropy method with rolling windows. We verify that this new method can capture the dynamic order between sequences and better reveal the nonlinear causality between nonstationary time series. Furthermore, we construct an investor sentiment index based on principal component analysis, and based on the proposed dynamic transfer entropy model, we analyze the information transfer relationship among economic policy uncertainty (EPU), investor sentiment and stock markets. The results of the information flow analysis of EPU and investor sentiment show that EPU influenced investor sentiment mainly from August 2015 to June 2016. Among different policies, China’s exchange rate reform policy and ‘circuit-breaker’ policy in the stock market have played an important role. The analysis of the information flow between sentiment and stock price returns shows that investor sentiment is more a reflection of changes in stock price returns with a 1-month lag order and that the stock market has a significant bargainer effect and a weaker bandwagon effect. There is no significant information flow transmission relationship between EPU and stock market volatility, which indicates that stock market fluctuations are basically not affected by national policy fluctuations. Although investor sentiment is affected by changes such as exchange rate reform and stock market policies, many investors do not form consensus expectations.