A Sustainable Quantitative Stock Selection Strategy Based on Dynamic Factor Adjustment

Abstract

In this paper, we consider a sustainable quantitative stock selection strategy using some machine learning techniques. In particular, we use a random forest model to dynamically select factors for the training set in each period to ensure that the factors that can be selected in each period are the optimal factors in the current period. At the same time, the classification probability prediction (CPP) of stock returns is performed. Historical back-testing using Chinese stock market data shows that the proposed CPP quantitative stock selection strategy performs better than the traditional machine learning stock selection methods, and it can outperform the market index over the same period in most back-testing periods. Moreover, this strategy is sustainable in all market conditions, such as a bull market, a bear market, or a volatile market.