Geographical information system based economic and financial risk analysis: The Case of Europe and Central Asia

The aim of the study; It is to classify each factor that constitutes the economic and financial risk determined by the ICRG rating agency by using a different classification technique and to compare them by creating thematic maps of the countries according to this classification. The other purpose is to create economic and financial risk maps of countries according to the economic and financial risk factors determined by the ICRG rating agency, using the geographically weighted regression method, which is one of the Geographical Information System (GIS) analysis techniques, in a way that will reveal dynamic and visual results. For this purpose, 10 factors included in economic and financial risk, which are among the country risk categories, were used according to the country risk model of the International Country Risk Guide (ICRG). GDP per capita, real GDP growth, inflation rate, budget balance as % of GDP, current account as % of GDP variables are used as factors determining economic risk. In the factors that determine the financial risk, foreign debt as % of GDP, foreign debt as % of exports of goods and services, current account as % of exports of goods and services, net international liquidity in months of import, exchange rate stability variables are used. In the study, 48 countries in Europe and Central Asia were determined to cover the years 2018 and 2019. In the study, using these parameters, economic and financial risk maps of 48 countries in Europe and Central Asia for the years 2018 and 2019 were created by geographically weighted regression (CAR) analysis. ArcGIS 10.8 program was used for CAR analysis. Before the CAR analysis of the variables included in the research, Moran's Index analysis was performed as the basic measurement of spatial autocorrelation and it was observed that the economic and financial risk dependent variables had statistically significant and positive autocorrelation, that is, spatial dependence, since the p-values of the dependent variables were less than 0,05 significance level. In addition, it was determined that the adjusted R2 values showing the performance of the established models were 0,999 in both 2018 and 2019, and the independent variables explained the dependent variables at a high rate of % 99,9. According to the results of the CAR analysis, the risk level of the countries examined was divided into five classes (very high risk, high risk, medium risk, low risk and very low risk), and risk maps were created according to the predicted values of the economic and financial risk scores obtained as a result of the CAR analysis, and more dynamic, meaningful and visual results were revealed. It has been seen that the CAR analysis results obtained in the study can be used in GIS-based applications in the field of economy and finance, and in this direction, a new interdisciplinary perspective has been brought to the literature in determining the economic and financial risk of countries.