Stock Price Prediction Model Integrating an Improved NSGA-III with Random Forest

Stock price prediction models have attracted much research interest in recent years. However, stock prices are high-dimensional financial time series. The application of artificial intelligent (AI) algorithms are widely used in stock price prediction models. How to improve model accuracy and reduce computing costs are the important indicators for prediction models. In response to the above issues, the prediction system based on the INSGA-III-RF algorithm established in this article can be regarded as a multi-objective feature selection problem. The INSGA-III-RF algorithm obtains the optimal feature subset through a heuristic search strategy. The solution selected by the algorithm has high classification accuracy and low number of features. The experimental results show that INSGA-III-RF model has better performance compared with benchmark models. INSGA-III-RF model has achieved the optimization goal of maximizing accuracy and minimizing the size of the solutions.