Comparing the univariate modeling techniques and box-jenkin for measuring of climate index in sitiawan, Malaysia

Abstract

The purpose of the article is to determine the most suitable technique to generate the forecast models using the data from the series of climate index in Sitiawan, Perak. This study are using univariate time series models and box-jenkin consists of  Naïve with Trend Model, Single Exponential Smoothing , Double Exponential Smoothing, Holt’s Method, Adaptive Response Rate Exponential Smoothing (ARRES), Holt-winter's Trend and Seasonality and SARIMA model. Using time-series data from 1961-2012 (monthly), there’s several data consists missing/outlier value. The issues are overcome with applied the time series model for each missing values and then compare the measure error (mean square error, MSE) for each models. Then, the selection of the most suitable model was indicated by the smallest value of mean square error, MSE. Based on the analysis, SARIMA(0,1,3)(0,1,2)12 model is the most suitable model for forecasting the climate index in Sitiawan, Perak.