Discrete Wavelet Transform-based feature engineering for stock market prediction

Stock market prediction is an interesting area of research where Technical Indicators (TI) play an important role. However, prediction of stock market movement is difficult due to the presence of noise and irregularities in the stock data. Data de-noising and decomposition techniques are apt to handle such noise. The data decomposition technique may lead to the generation of a large feature vector that needs to be handled carefully. Therefore, a suitable and effective feature engineering component must be included in the prediction model. To handle the above-mentioned issues, this paper proposes a stock market prediction model which offers a module for TI computation, feature engineering, and stock market prediction. A feature engineering component is proposed in which Discrete Wavelet Transform (DWT) is offered for data decomposition and Chicken Swarm Optimization (CSO) is offered to handle the large number of features generated through DWT. CSO is used to select the optimal feature subset. The proposed feature engineering component is named as DWT-CSO. The stock market trend prediction is performed by Machine Learning (ML) and Deep Learning (DL) models. The dataset of Indian (NIFTY50 and BSE) and US stock (S&P500 and DJI) indices is used for experimentation. The proposed DWT-CSO provided improved performance. The prediction models’ accuracy is increased by 19.59% (for S&P500), 18.33% (for DJI), 19.43% (for NIFTY50), 15.89% (for BSE). The performance of DWT-CSO is statistically analysed with Wilcoxon rank-sum test. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.