Analysis of Financial News Impact on Stock Based on a Statistical Learning Method with News Density

摘要

Since the investors often react to news and consequently make stock prices move, financial news has an impact on stock prices. However, the price adjustment process is a complex one. In this paper, a statistical learning methodology has been proposed to analyze the time lag between the price moves and the released news. We firstly set up a news-price mapping model to study the relationships among the price moves and the volume of the news articles. Cross validation has also been used to test the time lag. Experiments have been conducted by using the daily data sources in 2002 in the Hong Kong stock exchange market, and the results have showed that in most cases, the time lag is approximately equal to 1 day.