PEMODELAN ANTAR VARIABEL EKONOMI SECARA SIMULTAN MENGGUNAKAN PENDEKATAN VECTOR ERROR CORRECTION MODEL (VECM)

Abstract

The movement of the Jakarta Composite Index (IHSG) is influenced by internal factors such as inflation, the BI Rate, exchange rate, and external factors consisting of world gold prices and world crude oil prices. The six economic variables have a relationship simultaneously. Vector Error Correction Model (VECM) is a Vector Autoregressive (VAR) which has non-stationary but has a long-term cointegration. The purpose of this study is to analyze the cointegration among economic variables and determine the model of economic variables. Data for the variables is monthly data for the period January 2012 to December 2021 which has fulfilled stationarity at first level of difference. The optimum lag chosen is lag 1 so that the model to be used is VECM(1) and the resulting VAR system has less than one modulus for the VAR to be stable. Johansen's cointegration test yielded 5 cointegrations, so each short-term period adjusts simultaneously and tends to adjust with each other to achieve long-term equilibrium. The Mean Absolute Percentage Error (MAPE) value in the evaluation of model accuracy ranges below 10%, so the model’s performance is very good.