A Hybrid Intelligent Morphological Approach for Stock Market Forecasting

• Published in: Neural processing letters Vol. 31; no. 3; pp. 195 - 217
• Main Author: ARAUJO, Ricardo De A
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01.06.2010; Springer; Springer Nature B.V
• Subjects: Applied sciences; Artificial Intelligence; Back propagation networks; Complex Systems; Computational Intelligence; Computer Science; Computer science; control theory; systems; Connectionism. Neural networks; Exact sciences and technology; Forecasting; Genetic algorithms; Mathematical models; Morphology; Neural networks; Phase distortion; Prediction models; Representations; Securities markets; Statistical tests; Time series; Morphological neural networks; Hybrid intelligent models; Stock market forecasting; Genetic algorithms; Morphological neural network; Time series; Real time; Modeling; Forecasting; Search algorithm; Time lag; Minimum time; Statistical test; Backpropagation algorithm; Genetic algorithm; Behavioral analysis; Stock exchange; Hybrid model; Delay time; Financial data; Stock markets
• ISSN: 1370-4621; 1573-773X
• DOI: 10.1007/s11063-010-9133-1

In this paper, a hybrid intelligent morphological approach is presented for stock market forecasting. It consists of a hybrid intelligent model composed of a Modular Morphological Neural Network (MMNN) and a Modified Genetic Algorithm (MGA), which searches for the minimum number of time lags for a correct time series representation, as well as by the initial weights, architecture and number of modules of the MMNN. Each element of the MGA population is trained via Back Propagation (BP) algorithm to further improve the parameters supplied by the MGA. Initially, the proposed method chooses the most tuned prediction model for time series representation, then it performs a behavioral statistical test in the attempt to adjust time phase distortions that appear in financial time series. An experimental analysis is conducted with the proposed method using four real world time series and five well-known performance measurements, demonstrating consistent better performance of this kind of morphological system.