Entropy Based China Income Distributions and Inequality Measures

Abstract

We use entropy based methods on a sample of household income data from the 2005 Chinese Inter-Census as a basis for summarizing the income data in the form of a probability density function(PDF), and to obtain measures of income inequality. We estimate the income probability density functions using information theoretic entropy based divergence measures and recover the corresponding inequality measures for each of China’s provinces. In the entropy criterion-measure we seek a probability density function solution that is as close to a uniform PDF-distribution of income (an equal distribution with the least inequality), as the micro sample data will permit. These entropy measures reflect cross sectional measures of income inequality and reflect how the province level economic behavioral system is functioning and how the allocation and distribution systems is performing. Finally we use a sample of data from the China Family Panel Study to recover an estimate of the income PDF for China in 2016.