A Novel Model for Predicting Stock Index Trends through Hybrid Observed Mode Decomposition-Based Optimized Dynamic Sequential Extreme Learning Machine
Sumathi R, Sigamani Ashokkumar · International Journal of Electronics and Communication Engineering · 2024
In the dynamic field of financial markets, the precise prediction of Stock Index Trends (SIT) has long been a significant objective for investors and traders. As global markets evolve, conventional models encounter challenges in keeping abreast of the intricacies and rapid transformations inherent in today’s financial ecosystems. The innovative SI Trend Prediction (SITP) model introduced in this study integrates a novel method by integrating the Hybrid algorithm with Observed Mode Decomposition (OMD) and Optimized Dynamic Sequential Extreme Learning Machine (ODS-ELM), denoted as OMD-ODSELM. The Hybrid IHS algorithm is deployed to optimize the model parameters, thereby enhancing the efficiency and convergence nothe decomposed data is input into ODS-ELM, a Machine Learning (ML) algorithm suitable for online learning scenarios, to predict real-time SITs. This hybrid model capitalizes on the IHS algorithm’s enhanced optimization capabilities and leverages the strengths of OMD and ODS-ELM for robust and accurate Stock Market Trend Prediction (SMTP). Overall, it is a valuable tool for investors and financial analysts in decision-making. The proposed model significantly contributes to financial prediction by providing a robust and efficient tool for predicting SITs, facilitating informed decision-making for investors and financial analysts.