Analisis Perbandingan Metode Granger Causality VECM, Granger Causality-ARDL dan Toda Yamamoto Causality dalam Memeriksa Hubungan Kausalitas antar Parameter Perekonomian Indonesia

Abstract

Economic globalization in recent decades has significantly changed the structure and dynamics of national and international economies. These changes can be seen from economic variables such as gross domestic product, inflation, unemployment, exports, imports, and money supply. A deep understanding of these dynamics is crucial for policymakers, businesses, and the general public to respond wisely to global economic changes. One important aspect to understand is the causal relationship between exports, imports, money supply, and inflation. This Causality analysis helps identify the extent to which one variable can influence another. Causality test methods used include Granger Causality in the VECM model, ARDL, and Toda Yamamoto. In VECM, the Granger Causality relationship is modeled through an error correction term. ARDL tests and models the long-term dependentcy of a variable on its previous values. Toda Yamamoto tests the direction and significance of Causality between two or more variables in the VAR (Vector Autoregression) model. Performance analysis of these methods based on AIC values shows that ARDL has the best performance in identifying causal relationships among economic variables.