A Cross-Correlation-Based Stock Forecasting Model

Sungil Kim


Researchers are continuously seeking to develop and improve the stock forecasting models by analyzing the past value of a company and predicting future performance based on past data trends. Many previous studies of technical analyses of stocks have focused heavily on forecasting a single stock price based on its own past data. This type of analysis is susceptible to stock market volatility and not very effective for intraday trading. In addition, it is difficult to apply a method used for a particular stock market sector to other market sectors. In this study, we present a cross-correlation-based forecasting model using sets of closely related stocks to forecast future stock performance. For highly correlated pairs with a time delay of K days, the two stocks are assumed to exhibit a similar pattern in the short term—that is, predicting stock B’s price based on stock A’s price will reflect stock A’s future performance K days earlier. The forecasting model generates a buy or sell signal depending on the performance of one stock that influences the other stock with the lag. The accuracy of the developed model is measured using US stocks from the energy sector, which is more volatile than other indexes (i.e., S&P 500), and the technology sector. The proposed model accurately forecasts outcomes 87.2% of the time and generates 3.2% profit per dollar over the span of the 47-day forecast interval. This result shows that the developed forecasting model is ideal for high-risk, high-return investments.


Stock Market, Cross-Correlation, Forecasting

Full Text: PDF


  • There are currently no refbacks.