Panel Regression Models for Measuring Multidimensional Poverty Dynamics

This work is related to the studying of poverty dynamics and the analysis of the socio-demographic factors influencing it. We have chosen a fuzzy and multidimmensional approach in order to define two different poverty measures. A panel regression model has been estimated and particular attention has been paid to the treatment of the unobservable heterogeneity among longitudial units. The specified model combine autoregression with variance components. The empirical analysis has been conducted using the data set of the British Household Panel Survey from 1991 to 1997.