Investor sentiment and market return predictability: evidence from the Stock Exchange of Thailand

Investor sentiment is the emotions of investors which affect investors’ decisions and bias. It is an interesting issue whose effects many researchers have tried to investigate, especially in relation to stock markets. This paper studies the impact of sentimental factors on the Stock Exchange of Thailand (SET) by using sentiment-proxy variables (number of IPOs, average first-day return of IPOs, market turnover, net added new account and market performance) to forecast Thai stock market returns in the SET and MAI markets. A rolling-window technique is applied to increase the predictive power of the model. The empirical results show that including sentiment-proxy variables in a forecasting model cannot help control variables (exchange rate, interest rate and inflation rate) to predict stock market returns accurately in either the SET or MAI market. However, a model which includes only sentiment-proxy variables can forecast MAI TRI more accurate than a comparison model but not give the same results for SET TRI. It indicates that the sentiment factor can affect small-cap stocks more than big-cap stocks in the case of the Stock Exchange of Thailand.