BIDIMENSIONAL STOCHASTIC DOMINANCE METHOD IN POVERTY RATE COMPARISON

A stochastic dominance is a method to compare the dominance of two distribution functions. In poverty, the stochastic dominance is used to compare poverty rates across regions and time. The poverty comparison using stochastic dominance is robust to the selection of poverty line. The bidimensional stochastic dominance is the stochastic dominance method in two dimensions. This research aims to determine the use of a first and a second order bidimensional stochastic dominances, derive the statistical test in the bidimensional stochastic dominance method and apply the method in poverty comparison of Central Java. Here, poverty rate is influenced by an average of expenditures per capita and years of schooling. The average of expenditures per capita is used as the first dimension and years of schooling as the second dimension. Poverty rate is seen from proportion of the poor for the first orde bidimensional stochastic dominances and the second order is seen from the average poverty gap. The statistical test in the bidimensional stochastic dominance method is the asymptotically standard normal distribution. The application of the method generates poverty rate seen from the average poverty gap of Central Java in 2010 higher than the average poverty gap in 2011.