Comparison of weighted Markov chain and fuzzy time series Markov chain in forecasting stock closing price of company X

Abstract

The stock is one of the most popular financial market instruments. For various purposes, many methods have been used to forecast stock price in the future. The purpose of this paper is to compare the Weighted Markov Chain (WMC) method and the Fuzzy Time Series Markov Chain (FTS-MC) method in forecasting the stock closing price in the future. The basic idea of both methods is to forecast the possibility of future stock prices based on the historical data of stock prices in the past. Nevertheless, both methods have a different approach in transforming the value of stock prices to the range of the random variables (the states). This paper considered the daily stock closing prices data of company X in Indonesia from January 2, 2017 to December 29, 2017. The accuracy of both methods is verified by using the Mean Absolute Percentage Error (MAPE). The result shows that the Fuzzy Time Series Markov Chain has smaller MAPE of 1.7404%.