Investigating the Factors Affecting Multidimensional Poverty Using Linear Mixed Beta Regression Model

Multidimensional poverty is a complex issue that is affected by various factors. Therefore, accurate knowledge and its comprehensive definition are great important. The aim of this study was to investigate the effect of economic, social and institutional factors on multidimensional poverty based on a comprehensive measurement framework and using the beta linear mixed regression model, which is a suitable for response variables in the range (0,1). The parameters are calculated with the Bayesian approach. The Multidimensional Poverty Index has been calculated using the Multiple-Indicator Demographic and Health Survey 2015 and the Alkier-Foster method. In the group of economic factors affecting multidimensional poverty, four variables of inflation rate, unemployment rate, economic participation rate and net rate of provincial immigrants, in the group of social factors, two variables, level of development and urbanization, and in the group of institutional factors, three variables of crimes and the number of people supported by social harms, and the level of satisfaction of the administrative apparatus are considered for modeling. The results showed that both economic and social factors affect the multidimensional poverty index. Inflation rate and urbanization rate affect the average and unemployment rate affect the variance of multidimensional poverty among provinces