One success comes from the technical analysis, which is a method of evaluating securities by analyzing the statistics generated by market activity. It captures the fact that market participants tend to provide a consistent reaction to similar market stimuli over time. Based on the assumption that all information has been discounted by the price and volume, technicians think that they can predict the price movement by identifying the price patterns that can suggest future activity. This means that after a trend has been established, the future price movement is more likely to be in the same direction as the trend than to be against it [ Technical Analysis: The Basic Assumptions (n.d).]. Using the estimated future price, technicians naturally decide whether the current price is low or high in the time domain. While technical analysis makes sense when stock price movement is only affected by single factor, it is not suitable to analyze the movement caused by multi-factors.
Another success comes from fundamental analysis, which considers that stock price movement is affected by many factors, which have not been discounted by the price and volume. It classifies the factors into market factors and fundamental factors, and thinks intrinsic value, determined by fundamental factors, is fluctuated by stock price. Fundamentalists can decide whether current price is low or high by selecting a proper safety margin [Benjamin Graham, and David Dodd. (1951)].
Recent success comes from text data analysis, which points out that the release of a financial news article can trigger the change of stock price.Falinouss, P. (2007) can predict the changes of stock trend in a time window on the basis of analyzing the information in the news article. Schumaker, P. R. ,and Chen, H go further. They could predict discrete stock price within a period of twenty minutes before and twenty minutes after the news release [Schumaker, P. R. , Chen, H. (2006)], capture short-term surges in price led by overreaction of investors to news [Schumaker, P. R. , Chen, H. (2008)], and have a 2% higher return than the best performing quant fund with their system AZFinText[Schumaker, P. R. , Chen, H. (2009)].