Stephen M. Miller (Business and Economic Research) and Heni Boubaker, Institute of High Commercial Studies (IHEC) of Sousse, Giorgio Canarella, (emeritus) University of California, Los Angeles, and Rangan Gupta, University of Pretoria, published: "" in Computational Economics on September 27, 2022.
This paper forecasts returns and volatilities of the stock market. The ARFIMA-WLLWNN model integrates the advantages of the ARFIMA model, the wavelet decomposition technique, and artificial neural network. A wavelet decomposition improves the forecasting accuracy of the LLWNN neural network, resulting in the Wavelet Local Linear Wavelet Neural Network (WLLWNN) model. The Back Propagation and Particle Swarm Optimization (PSO) learning algorithms optimize the WLLWNN structure. The residuals of an ARFIMA model of the conditional mean become the input to the WLLWNN model. The hybrid ARFIMA-WLLWNN model is evaluated using daily returns of the Dow Jones Industrial Average index from 01/05/2010 to 02/11/2020.
The experimental results indicate that the PSO-optimized version of the hybrid ARFIMA-WLLWNN outperforms the LLWNN, WLLWNN, ARFIMA-LLWNN, and the ARFIMA-HYAPARCH.
