The Influence of Investor Attention on Return and Volatility of Stock Market

Abstract

Applying internet data to investor attention research is a trend now. This paper propose a new method to measure positive and negative investor attention paid to a certain industry by using of search data from search engine. We select keywords from a corpus of energy industry by text-analysis technique including TextRank algorithm instead of taking company names or stock tickers as keywords in previous studies. After selecting keywords, we calculating two indices by PCA method. The empirical analysis demonstrates that the positive index has significantly negative effect on stock index volatility. However, the negative index has significant influence on volatility. These results shows the rationality of our indices construction.