| Abstract: | Forecasting stock market is an exciting topic. In this study, we would like to employ text mining tools and statistical modelling to forecast Chinese stock market movement based on collective viewpoints of individual investors. We quantify the collective viewpoint of individual investors by analysing and aggregating their online posts related to the stock market, where the main challenges -lie in how to automatically identify a set of suitable keywords. We collect three years of data from one of the largest social media networks Sina Weibo in China. Our proposed hybrid time series regression model produces more accurate forecasts than existing models. Our statistical results suggest that there exists a strong causality relation between the collective viewpoints reflected on Sina Weibo and the movement of stock market in China. |
| Date: | 2 October 2013 |
| Time: | 2:30pm - 3:00pm |
| Speaker: |
Mr Chen Yibo PhD Student Department of Management Sciences City University of Hong Kong |
| Venue: | Room Y5-412, 5/F, Academic 1, Yellow Zone, City University of Hong Kong |
[ Back ]
