Comparison of Fuzzy Time Series and ARIMA Model for Predicting Stock Prices

Abstract

The stock market has always been a contentious topic in society, and it is a place where economic standards are established. The stock market is incredibly unpredictable and turbulent. This means that the shares may fluctuate for reasons that are sometimes difficult to understand. Due to this uncertainty, many investors believe the stock market as a risky investment. Therefore, having an accurate picture of future market environment is crucial to minimising losses. Forecasting is a technique of predicting the future based on the outcome of the previous data. There are a wide range of forecasting algorithms, however, this study only focuses on these two techniques: Auto Regressive Moving Average (ARIMA) model and Fuzzy Time Series (FTS) Model. The goal of this study is to evaluate and compare the effectiveness of the ARIMA model and the FTS model in predicting sample data of stock prices of Top Glove Corporation Berhad since this company is the largest glove supplier in the world and plays a significant role in the Covid-19 global pandemic crisis. The error measures that were taken into consideration consist of Root Mean Square Error (RMSE), Mean Square Error (MSE), and Mean Absolute Percentage Error (MAPE). These measurements were computed numerically and graphically using a statistical programme called EViews. The outcome shows that the ARIMA model performs better than the FTS model in terms of forecasting accuracy and provides the lowest values of MAPE, MSE, and RMSE, which are 10.58757, 0.926354, and 0.962473, respectively.

Keywords

• Autoregressive Integrated Moving Average
• ARIMA
• Fuzzy Time Series
• Stock Price Forecasting
• Share Price Forecasting