Open AccessSELECTION OF INPUT VARIABLES OF NONLINEAR AUTOREGRESSIVE NEURAL NETWORK MODEL FOR TIME SERIES DATA FORECASTING
Author(s) -
Hermansah Hermansah,
Dedi Rosadi,
Abdurakhman Abdurakhman,
Herni Utami
Publication year - 2020
Publication title -
media statistika
Language(s) - English
Resource type - Journals
ISSN - 2477-0647
DOI - 10.14710/medstat.13.2.116-124
Subject(s) - exponential smoothing , autoregressive integrated moving average , autoregressive model , artificial neural network , computer science , time series , model selection , series (stratigraphy) , mean squared error , selection (genetic algorithm) , nonlinear system , nonlinear autoregressive exogenous model , smoothing , artificial intelligence , machine learning , data mining , statistics , mathematics , quantum mechanics , computer vision , biology , paleontology , physics
NARNN is a type of ANN model consisting of a limited number of parameters and widely used for various applications. This study aims to determine the appropriate NARNN model, for the selection of input variables of nonlinear autoregressive neural network model for time series data forecasting, using the stepwise method. Furthermore, the study determines the optimal number of neurons in the hidden layer, using a trial and error method for some architecture. The NARNN model is combined in three parts, namely the learning method, the activation function, and the ensemble operator, to get the best single model. Its application in this study was conducted on real data, such as the interest rate of Bank Indonesia. The comparison results of MASE, RMSE, and MAPE values with ARIMA and Exponential Smoothing models shows that the NARNN is the best model used to effectively improve forecasting accuracy.