A comparison of fuzzy time series with statistical analyses in forecasting Malaysian goverment tax revenue

Abstract

Many forecasting models based on the concept of fuzzy time series have been proposed in the past decades. In recent years, many researchers have used fuzzy time series to handle forecasting various domain problems and it has been shown to forecast better than other models such as the predictions of stock prices, academic enrollments, weather, road accident casualties, etc. However, two main factors, which are the lengths of intervals and the content of forecast rules, impact the forecasted accuracy of the models. This paper presents a simple fuzzy set theory and fuzzy time series forecasting method of order three towards Malaysian government tax revenue which uses a time variant difference parameter on current state to forecast the next state. Based on the relationship, the forecast of the government tax revenues is generated in fuzzy terms, such as: 'moderate value', 'poor value', 'excellent value' and etc. The accuracy of using the different number of fuzzy sets on the prediction of the government tax revenue has shown and compared in this paper.