Comparative Analysis of ARIMA, Fuzzy Time Series Method and Hidden Markov Model for Stock Market Prediction

Stock market forecasting is a challenging task for the researchers. Many statistical and machine learning methods with varying degree have been developed to test the accuracy of forecasting. The main purpose of this paper is to compare the forecasting accuracy of ARIMA, Fuzzy Time Series (FTS) and Hidden Markov Model (HMM). Our analysis of performance measure is based on Mean Error, Mean Square Error, Mean Absolute Deviation and Mean Absolute Percentage Error. Experimental result showed that Fuzzy Time Series with technical indicators achieved better forecasting accuracy than ARIMA and Hidden Markov Model.