Model nonlinear in economic and finance riset are often found. One model that can be used to capture nonlinear relationships in data is Threshold Vector Autoregressive (TVAR) model. TVAR model is generalization of VAR model, it divides the time series into different regimes that are separated by a different threshold. The purpose of this research are is to see the effects between inflation, exchange rate depreciation and money supply in Indonesia and to know the performance of forecasting with VAR and TVAR model. Inflation, exchange rate depreciation and money supply growth produce TVAR on lag 1 with one threshold and two regimes. Each regimes shows different effects. TVAR model has a smaller AIC than the VAR model so forecasting performance of TVAR model is better used in this modeling.
展开▼
