NON LINEAR AUTOREGRESSIVE MOVING AVARAGE UNTUK PREDIKSI KUNJUNGAN WISATAWAN
Keywords:
Prediction, Time Series, Nonlinear, Autoregressive, Moving Average, Narma
Abstract
Prediction is one of the elements for decision support in the future. Because of the unavailability of natural resources such as oil and gas, forest products or large-scale manufacturing industries in Lombok, tourism has become a leading sector in economics development. Prediction of tourist arrivals needs to be done to support policies related to tourism development. There are two basic methods of prediction: arima and neural network. Arima is good for prediction with stationary dataset, while neural networks are either used for prediction with stationary or non-stationary data. Previous research related to the prediction of tourist arrivals using Recurrent Neural Network approach with Extended Kalman Filter. This research tries to predict time series data from tourist arrivals by Nonlinear Autoregressive Moving Average (NARMA) approach. Predicted results based on Mean Square Error (MSE), the best prediction result is given on ARMA model (5,1,0) with MSE 0.178.
References
[1] A. A. Rizal and S. Hartati, "Recurrent neural network with Extended Kalman Filter for prediction of the number of tourist arrival in Lombok," 2016 International Conference on Informatics and Computing (ICIC), Mataram, 2016, pp. 180-185.
[2] Brath, A., Castellarin, A., & Montanari, A. 1999. Detecting non stationarity in extreme rainfall data observed in Northern Italy. In Proceedings of EGS-Plinius Conference on Mediterranean Storms, Maratea (pp. 219-231).
[3] Mulyono, S. 2006. Statistik untuk Ekonomi dan Bisnis Edisi Ketiga. Jakarta: Lembaga Penerbit Fakultas Ekonomi UI.
[4] Makridakis, S. dan Steven, W., 1993, Metode dan Aplikasi Peramalan, Jakarta : Penerbit, Erlangga.
[5] Adnyani, L.P.W., 2012, General Regression Neural Network (GRNN) Pada Peramalan Data Time Series, Tesis, S2 Matematika UGM, Yogyakarta.
[6] Munarsih, E., 2011, Penerapan Model Arima-Neural Network Hybrid Untuk Peramalan Data Time Series, Tesis, Program Studi S2 Matematika, Universitas Gadjah Mada, Yogyakarta.
[2] Brath, A., Castellarin, A., & Montanari, A. 1999. Detecting non stationarity in extreme rainfall data observed in Northern Italy. In Proceedings of EGS-Plinius Conference on Mediterranean Storms, Maratea (pp. 219-231).
[3] Mulyono, S. 2006. Statistik untuk Ekonomi dan Bisnis Edisi Ketiga. Jakarta: Lembaga Penerbit Fakultas Ekonomi UI.
[4] Makridakis, S. dan Steven, W., 1993, Metode dan Aplikasi Peramalan, Jakarta : Penerbit, Erlangga.
[5] Adnyani, L.P.W., 2012, General Regression Neural Network (GRNN) Pada Peramalan Data Time Series, Tesis, S2 Matematika UGM, Yogyakarta.
[6] Munarsih, E., 2011, Penerapan Model Arima-Neural Network Hybrid Untuk Peramalan Data Time Series, Tesis, Program Studi S2 Matematika, Universitas Gadjah Mada, Yogyakarta.
