Comparative analysis of Fuzzy Regression and Fuzzy ARIMA models for forecasting real gross domestic product of Nigeria

Abstract

Over the years, a number of studies have been conducted in terms of forecasting the real gross domestic product (GDP) of Nigeria. The GDP growth rate measures the annual growth rate in percentage of the monetary value of all finished goods and services made within a country. It is an important indicator of economic growth. This paper presents Fuzzy ARIMA and Regression to determine the interval of possibility for predicting the real GDP. The Fuzzy ARIMA (FARIMA) and Fuzzy Regression (FR) methods requires small size data as compared to the classical time series. Comparison between FARIMA and Fuzzy Regression with a threshold level of zero (ℎ=0) is performed by calibrating the models on existing data. The minimum values of the total spreads for FARIMA and FR are 0.791 and 4.077 respectively. In addition, the MAPE values for FARIMA is smaller than that of FR. Furthermore, the results indicate that the FARIMA gives a narrower interval of possibility for prediction than the FR.