Predicting Stock Indices Trends using Neuro-fuzzy Systems in COVID-19

Abstract

Predicting the ebb and flow of stock markets is a complex and challenging exercise owing to the disruptive and uncertain behavior of stock prices. The COVID-19 pandemic is an example of an event that, had a drastic impact on global stock markets, due to business activities and trading being severely affected. It is important, therefore, to be able to predict how stock markets behave in a crisis period. We find that stock markets obtain the worst returns in countries where there are higher reported positive cases of coronavirus. This study employs adaptive neuro-fuzzy inference systems (ANFIS), comprising of a controller and the stock market process, to predict the behavior of selected stock indices. After training ANFIS and evaluating the resultant data, we estimate statistical errors and found that 100 training epochs provide marginally better results. To test the accuracy of our results, we used hit rate success and report that the neuro-fuzzy system predicts stock market trends with an average accuracy of 65.84%, an improvement over earlier techniques reported in the literature. Finally, we compute the rate of return using a buy-and-hold strategy and a neuro-fuzzy system, and identify that market indices outperform by employing the proposed method.