Predicting Stock Market Returns in Turbulence through the Adaptive Neuro-Fuzzy Inference System

Abstract

Accurate stock market forecasting impacts the decision-making of firms, investors, and macroeconomic policies. This study analyzes the performance of neuro fuzzy technique for stock market prediction in turbulent and normal states for firms listed on the Pakistan Stock Exchange. The purpose of this study is to examine the forecasting accuracy of neuro fuzzy. It further considers the ex-ante uncertainty and information asymmetry theories. To determine one-day ahead forecasting, we consider the dataset of 15 non-financial firms for two volatile market periods (2017 market fluctuations and covid-19) and one stable market period (2018-2019). Two adaptive neurofuzzy inference systems (ANFIS) are employed in the methodology; the controller and the stock market process. Results suggest an accurate prediction of stock market trends using the proposed system in all three market periods. They were evaluated based on hit rate percentage, root mean square error (RMSE) and graphical representation. Results from the controller model were used to drive the stock market process model, whereas results from the process model indicate that stock market volatility, change in stock prices and previous day stock returns can effectively predict next day stock returns using ANFIS. The findings would be useful for investors in yielding higher returns during stable well as turbulent market periods. Furthermore, it would aid managers, policy makers and analysts in decision making and target as setting.