An Enhanced Hybrid Model for financial market and economic analysis: a case study of the Nasdaq Index

Abstract

Individuals participate in the purchase and sale of securities affiliated with corporations on the stock market, which increases economic prosperity. The intricate interplay between economic factors, market dynamics, and investor psychology poses a significant challenge in accurately predicting outcomes within the field of finance. Additionally, the presence of non-stationarity, non-linearity, and high volatility in stock price time series data exacerbates the challenge of making precise estimations about stock prices in the securities market. The use of conventional techniques has the capacity to augment the accuracy of predictive modeling. However, it is important to acknowledge that these approaches also include computational intricacies, which might result in a higher likelihood of errors in predicting. This research introduces a novel model that adeptly addresses several issues via the integration of the Ant lion optimization methodology with the radial basis function method. The hybrid model showed greater effectiveness and performance in comparison to other models in the current study. The proposed model demonstrated a significant degree of effectiveness, characterized by optimum performance. The usefulness of a proposed predictive model for projecting stock prices was assessed by an analysis of data obtained from the Nasdaq index. The data covered the time period from January 1, 2015, to June 29, 2023. The findings suggest that the suggested model demonstrates reliability and effectiveness in its ability to analyze and predict the time series of stock prices. The empirical results suggest that the suggested model has a higher level of predictive accuracy in comparison to the other approaches by having the highest value of 0.991 for the coefficient of determination.