Stock Selection through Hidden Markov Model: A Case of Pakistan Stock Exchange

Abstract

To foresee hidden regimes inside the data, Hidden Markov Model is widely used. Many researchers had used various data mining methods to predict stock market prices. This research will describe usage of Hidden Markov Model (HMM) via MATLAB to forecast stock prices for stock’ selection and portfolios development. In this study listed companies of PSX (Pakistan Stock Exchange) are explored specifically KSE-100 Index companies. From January 2012 to June 2022 monthly closing stock prices are used in this study. Many studies based on the historical data have been conducted in developed markets, but we managed the contextual study of the Pakistan Stock Exchange to recognize the stream and variations happening for individual stocks in stock market of Pakistan. Outcomes of the research suggested a strong relationship between stock prices as projected by the Model.