Bayesian Spatial Econometric Analyses for the Growth of GDP and Regional Agglomeration Effects

This paper investigates the relationship between the growth of GDP and investment,consumption,and exportation by exploiting a series of spatial econometric models from a Bayesian perspective.The factor of regional agglomeration is also inducted into the model.The results show that the spatial correlation between economic growth in China is caused by external shocks,which behaves as spatial errors autocorrelation.By including the spatial correlation and spatial heterogeneity into the model,it's found that the contribution of consumption to economic growth plays the dominant role,and its impact exceeds the total effects of invest and exportation,which differs significantly from the results derived from the ordinary regression model.Meanwhile,the spatial econometric model of GDP growth exhibits regional agglomeration effects difference,which is shown by the fact that the growth of the western region is significantly below the other areas,while there do not exist distinct difference between the eastern and the middle areas.