Correlating Stock Index Movement with Investor Sentiment on Social Network
This work is supported by Chao-qiang Huang.
The investor sentiment hypothesis in behavioral finance tells us that emotions can cause to their decision-making. And social network has become a popular venue for sharing opinions. Here we want to discover whether there is a correlation between the sentiment index derived from Sina microblog and the 5-minute closing value of Shanghai Composite Index(SCI) in one day. In this paper, we put forward an efficient way to gather investor information and construct stock-oriented sentiment lexicons that measure positive vs. negative mood. Specifically, we got everyday blogs from many stock investor’s microblog using Sina API and designed an approach to the quantification of the text comments. In order to further analysis, we used three trading day’s real-time data to see the effects of our sentiment index computation. On the basis of two sequences, we tried to use Pearson Correlation to make analysis at different level. Then we applied our method to an intra-day market scale of three days to find that the max coefficient achieved 0.73 which means significant correlation and the minimum also got 0.56 that shows moderate relationship. Our results indicate that investor sentiment on social network has a positive correlation with SCI movement.
