Akademik Veri Yönetim Sistemi
WSEAS Transactions on Mathematics, cilt.4, sa.4, ss.368-373, 2005 (Scopus)
The analyses of spatial dependence and heterogeneity as well as issues on model selection have been predominant in spatial econometric modeling. This paper applies spatial econometrics to analyze the spatial dimensions of poverty and its determinants using data from Bangladesh. It is shown both theoretically and empirically that the parameter estimates of a poverty function estimated using ordinary least squares (OLS) suffer from upward bias. The alternative spatial econometric models showed that both spatial lag and spatial error parameters which are added in the poverty function are statistically significant, indicating that the OLS model is misspecified and that spatial effects are important in modeling the spatial dimensions of poverty. A comparison with the results from other studies is also provided. The paper concludes by drawing some implications and recommendation for further research.