LONG-RUN AND SHORT-RUN CAUSALITY BETWEEN STOCK PRICE INDICES AND MACROECONOMIC VARIABLES: EVIDENCE OF PANEL VECM ANALYSIS FROM BOSNIA AND HERZEGOVINA, CROATIA, NORTH MACEDONIA AND SERBIA

Abstract

The purpose of this paper is to identify the long-run and short-run relationship between the values of the Macedonian Stock Exchange Index composed of 10 most liquid listed stocks (MBI10), the Zagreb Stock Exchange Index (CROBEX) composed of the most liquid listed stocks, the Sarajevo Stock Exchange Index (SASX-10) composed of 10 most liquid listed stocks and the Belgrade Stock Exchange Index composed of 15 most liquid listed stocks (BELEX 15) and the selected macroeconomic variables. In order to identify the macroeconomic variables that affect the values of the selected stock indices, the analytical-synthetic method and the statistical method are applied. The statistical method uses econometric models for data analysis and interpretation and includes the application of the following econometric tools: Panel unit root test, Fisher -Johansen cointegration test, application of the panel vector error correction model (PVECM) and the Wald test statistics. The results of PVECM between the values of the selected stock indices and independent variables such as industrial production index 2015=100, average monthly gross wages, shows existence of conditionality or causal relationship on the long-run, when independent variable Harmonized Index of Consumer Prices (HICP) according to the COICOP classification 2015=100 is exluded from the model. By applying PVECM, it can be concluded that there is a long run causality running from independent variable to dependent variable, meaning that between the values of the selected stock indices and industrial production and average gross wages there is speed of adjustment towards long run equilibrium.