Estimating the macroeconomic ındicators using ARIMA and ANFIS methods

Creative Commons License

Kuzu Y. E., Alp S.

Recent Advances in Science and Engineering, vol.2, no.1, pp.6-17, 2022 (Peer-Reviewed Journal)


In this paper, inflation rates were predicted by using the adaptive neuro fuzzy inference system (ANFIS) and auto regressive integrated moving average (ARIMA) method. This study was carried out to contribute to the inflation forecasting studies in the literature and to diversify the forecasting studies made with artificial neural networks and traditional forecasting methods for the time series. Variables consisted of money supply, exchange rates and interest rates used in ANFIS model has been chosen by a detailed literature review. The data of this article were obtained from the Central Bank of Republic of Turkey. Results obtained from established models and the real values were compared using the performance criteria of root mean square error (RMSE), coefficient of  determination (R2), mean absolute error (MAE) and symetric mean absolute percentage error (SMAPE). Result illustrates the success of the ANFIS to predict the inflation rate and ANFIS model outperforms ARIMA model.