PERAMALAN INDEKS HARGA KONSUMEN MENGGUNAKAN MODEL INTERVENSI FUNGSI STEP

Abstract

Intervention model is a model for time series data in which practically there is an extreme fluctuation, whether it’s anupward or downward fluctuation. Consumer price index is one of economic data which plot has a fluctuation, the data that will being used for analyze is consumer price index of Indonesia in January 2009 until March 2015, on data detectable downward fluctuation significantly on January 2014 (T=61). Intervention in data was occurred in long time period (T=61 until T=75), so the model of intervention’s assumption is step function. Based on the result and analysis, the obtaining best model of intervention is ARIMA (2,1,3) with intervention order b=0 s=15 and r=0 which later on being used for predicting Indonesian consumer price index in six periods ahead. Keywords: consumer price index, stationery, ARIMA, step function intervention analysis, forecasting