A Stock Market Forecasting Model in Peru Using Artificial Intelligence and Computational Optimization Tools

Abstract

It is proposed the development of a forecast model capable of predicting the behavior of the price indices and quotes of the shares traded on the Lima Stock Exchange, based on the use of artificial intelligence techniques such as artificial neural networks and fuzzy logic based on computational optimization methods. The proposed model considers the forecast, in addition to the historical quantitative data of the share price, the inclusion of qualitative macroeconomic factors that significantly influence the behavior of the time series of the stock markets. It is about harnessing the ability of artificial neural networks to work with nonlinear quantitative data and their capacity for learning and also take advantage of the fuzzy logic technique to simulate the way of reasoning of human beings by defining judgment rules or knowledge base and their evaluation through inference mechanisms. The main contribution is to demonstrate that the proposed model is capable of obtaining more optimal approximations in the forecast of the financial time series. © 2020 Elsevier B.V., All rights reserved.