Penerapan Model Self-Exciting Threshold Autoregressive (SETAR) Nonlinear dalam Memodelkan Data Harga Minyak Sawit (FCPOc1)
DOI:
https://doi.org/10.22437/msa.v4i1.28292Keywords:
Forecasting, Nonlinear, Palm oil, SETAR, Time seriesAbstract
Palm oil is an agricultural commodity that has an important role in the global economy. Palm oil is obtained from the fruit of the oil palm tree (Elaeis guineensis) which grows in tropical regions, especially in countries such as Indonesia, Malaysia, Thailand and several West African countries. Palm oil has a variety of uses in the food, cosmetics and fuel industries, making it one of the most traded commodities in the world. Palm oil price fluctuations have a significant influence on the economy in producing and consuming countries. Therefore, a time series analysis is needed that can predict fluctuations caused by certain conditions. This analysis is carrying out nonlinear analysis using the SETAR (Self Exciting Threshold Autoregressive) method on palm oil prices to obtain a prediction model and prediction results for palm oil prices. The SETAR model is a special case part of the Threshold Autoregressive (TAR) model. The SETAR model threshold is a lag value of the series itself or the endogenous variable. Analysis carried out using the SETAR method produces a SETAR (3,1,1) model with threshold (r) = 0.01626070 where the fit value approaches the actual data value and the predicted value follows the actual data pattern
Downloads
References
Reeves, J. B., Weihrauch, J. L. 1979. Composition of foods: fats and oils. Washington, D.C.: U.S. Dept. of Agriculture, Science and Education Administration.
Tong, H. 1990. Non-Linear Time Series: A Dynamical System Approach. Oxford University Press. Oxford.
Wei, W. W. S. 2006. Time Series Analisis Univariate and Multivariate Methods, Second Edition. Pearson Education, Inc. New York.
Ghosh, H., Prajnesu dan Sunilkumar, G. 2006. Self,Exciting Threshold Autoregressive Models for Describing Cyclical Data. Calcutta Statistical association Bulettin, 58(1-2), 115-132. https://doi.org/10.1177/0008068320060108
Jaras, J., & A. M. Gishani. 2010. Threshold Detection in Autoregressive Non-linier Model. Departement of Statistics. Und University.
Makridakis, S., S. C. Wheelwright dan V. E. McGee. 1993. Metode Aplikasi Peramalan, Edisi Kedua. Alih bahasa Untung S. A. dan Abdul B. Erlangga. Jakarta.
Terasvirta, T., Lin, C. F. dan Granger, C. W. J. 1993. Power of the Neural Networks Linearity test, Journal of Time Series Analysis, 14(2), 209-220. https://doi.org/10.1111/j.1467-9892.1993.tb00139.x
Subanar & Suhartono. 2004. Uji Linieritas Tipe Lagrange Multiplier dengan Ekspansi Taylor untuk Deteksi Hubungan Nonlinier pada Data Time Series. J. Indones. Math. Soc. (MIHMMI), 1-15.
Hansen. 1999. Testing for linearity. Journal of Economic Surveys. 13(5): 551-576. https://doi.org/10.1111/1467-6419.00098
Ghozali, I. 2006. Aplikasi Analisis Multivariate dengan Program SPSS. Semarang: Universitas Diponegoro.
