A Principal Component Approach to Measuring Investor Sentiment in China
Haiqiang Chen
The Wang Yanan Institute for Studies in Economics, Xiamen University; MOE Key Laboratory in Econometrics, Xiamen Univeristy
Terence Tai Leung Chong
Yingni She
12/1/2013 8:32:37 PM
Abstract:
This paper develops a new investor sentiment index for the Chinese stock market. The index is constructed via the principal component approach (PCA), taking six important economic and market factors into consideration. The sentiment index serves as a threshold variable in a threshold autoregressive model to identify the stock market regimes. Our findings show that the Chinese stock market can be divided into three regimes: namely, a high-return volatile regime, a low-return stable regime and a neutral regime. The sentiment index is shown to have good out-of-sample predictability.
Keyword: 
Principal Component Analysis; Market Sentiment; Market Turnover; Threshold Model.