Informativeness of the market news sentiment in the Taiwan stock market

Abstract

This study constructs a market “aggregate news sentiment index” (ANSI) based on Chinese financial news relating to all the listed firms on the Taiwan Stock Exchange. We analyze the relationship between the ANSI and the market responses, which includes the returns, trading values, turnover ratios and volatility index, and the ANSI is divided into five regions. Our empirical results reveal that with an increase in the ANSI in the previous month, there is a corresponding increase in the trading value along with a reduction in the investor fear gauge of the Taiwan volatility index (TVIX). Granger causality tests indicate that both the weekly and monthly ANSIs could be leading indicators of market returns if the different regions of the ANSI are accounted for. In a potential application of our findings to portfolio management, we find that when the lagged term of the ANSI is classified in the highest regions, value-weighted portfolios comprising stocks with the highest market values, lowest price-to-book ratios and lower turnover ratios will earn the highest returns. The efficiency of the news sentiment in portfolio applications is consistent over time. This study contributes to the existing literature by demonstrating that the news sentiment level reflected in qualitative news reports can be effectively incorporated as a proxy to provide valuable support to portfolio decision making.