An Effective Stock Market Direction Using Hybrid WWO-MKELM technique

Abstract

Stock market return forecasting is currently regarded as a prediction issue. The forecasting process is challenging due to the financial markets inherent volatility on a global scale. The risks associated with investment procedures would be significantly reduced by the decrease in prediction error rate. To anticipate stock market return, this research offers a new hybrid WWO-MKELM technique. The three main processes of the described WWO-MKELM model are preprocessing feature extraction, and classification. First, the exponential smoothing approach is used to do preprocessing. The preprocessed dataset will then be used to extract the features. After that, a WWO-MKELM-based model is used to forecast stock prices. The WWO-MKELM model that has been described can foretell whether stock prices will increase or decrease. Utilizing the stocks of APPL and FB simulates the WWO-MKELM method. The obtained experimental findings showed that the WWO-MKELM model performed better than the compared approaches.