The Cointegration analysis of per capita trade turnover between the Republic of Azerbaijan and Turkey and the GDPs of these countries

The article builds an ECM model (error correction model) of the relationship between Azerbaijan's trade turnover with Turkey per capita and the GDP per capita of these countries for the period 1992–2022. The presented indicators reflect the level of activity and living standards in these countries. The analyzed time series are non-stationary in terms of their levels, and their first-order differences are stationary. All time series are logarithmized. The article uses the econometric methodology of gravity modeling of the relationship between the non-stationary time series. Various methods were correctly used during the modeling, including the extended Dickey-Fuller unit root test, Granger causality test, Engle-Johansen cointegration tests, vector error correction model, and standard diagnostic tests. Stationarity, causality, and cointegration tests were conducted on the entire sample at a significance level of 10 %. The existence of a statistically significant cointegration dependence of the balanced long-term relationship between the analyzed indicators is substantiated.