Application of ARCH model on nutmeg price forecasting in South Aceh district

The price of nutmeg which is changing and unstable in South Aceh results to volatility. Volatility is a variance pattern of time series, especially the time series of finances caused by variance not constant. This results the possibility of heteroscedastic data, so that it needs to be made a model of a particular approach to measure residual volatility problems. The data used in this research is the price of nutmeg in South Aceh district from January 2012 to December 2016 which has volatility. ARCH (Autoregressive Conditionals Heteroscedastic) is a model used to resolve the residuals variance which is not constant in financial time series. Then this model was developed into a Generalized Autoregressive Conditional Heteroskedastic (GARCH) to avoid too high orders on ARCH models and make the variance is always positive. The purpose of this research is to get the best ARCH/GARCH model for nutmeg price in South Aceh district and get the nutmeg price forecast for January 2016 until December 2017. The results of this research show that price of nutmeg in South Aceh district is stationary to the mean after the second differencing and the result of the ARCH Lagrange Multiplier test shows the problem in the 4 lag so the data are modelled with ARCH. The best models for forecasting price of nutmeg in South Aceh district is using model of ARIMA (2,2,0)-ARCH (4). While the results of the verification of the model are obtained the value of the MAPE of 6.12 percent.