Suppose you had data on stock price indices (X and Y) in two different countries. You are interested in causality issues and want to find out whether movements in stock prices in one country affect stock prices in another. If both variables are non-stationary and not cointegrated you can't model the dynamic relation between indices. You would test for cointegration if both prices are stationary. In case you find cointegration you would use a VAR model. O You would test for cointegration if both prices are non-stationary. In case you find cointegration you would use a VEC model. If both variables are stationary you would use a VEC model

Suppose you had data on stock price indices ( X and Y ) in two different countries. You are interested in causality issues and want to find out whether movements in stock prices in one country affect stock prices in another. If both variables are non-stationary and not cointegrated you can't model the dynamic relation between indices. You would test for cointegration if both prices are stationary. In case you find cointegration you would use a VAR model. You would test for cointegration if both prices are non-stationary. In case you find cointegration you would use a VEC model. If both variables are stationary you would use a VEC model