Spatial econometric models have become increasingly popular in economic studies, especially in recent years. These models differ from classical econometric models in that they take into account the effects arising from the location of the data set subject to the study. The spatial effect seen in a data set comes in two forms. The first form is spatial autocorrelation, which is defined as the correlation between neighboring locations. The other form is spatial heterogeneity, defined as the variance of a variable under consideration from place to place. The aim of this study was to find out whether the factors affecting provincial migration in Turkey have spatial characteristics. For this purpose, data from 2014 were selected as all data were accessible. A coherence matrix was constructed by considering the borders of the provinces to show the possible spatial relationship and this matrix was used in the econometric models. The distribution map of migration by province was examined and it was found that there was significant clustering, especially towards the northeast-southwest. From this point of view, the provinces where the spatial effect was significant in the clustered regions were identified using LISA statistics. The spatial model was determined in accordance with the orientation of the factors affecting migration and the movement of these variables was evaluated based on the provinces with significant correlation. The results obtained fit with the economic theory.
Migration Spatial Econometrics Spatial Autocorrelation Spatial Autoregressive Model
Birincil Dil | İngilizce |
---|---|
Bölüm | Makaleler |
Yazarlar | |
Yayımlanma Tarihi | 29 Haziran 2022 |
Gönderilme Tarihi | 15 Eylül 2021 |
Yayımlandığı Sayı | Yıl 2022 Sayı: 36 |