Scholarlink

<p>This study conducts a comparative analysis of quantile-based income distributions across Indian states, addressing a gap in the literature. It focuses on parametric models with closed-form quantile functions to model per capita household income data sourced from the India Human Development Survey II, excluding observations with negative or zero income and outliers. The Weibull, Power-Pareto, Singh-Maddala, Dagum, Singh Maddala Dagum, and Modified Lambda family were evaluated, with parameters estimated using the L-moments method. L-moments are advantageous because they are less affected by sampling variability and are robust to outliers. Model adequacy was first assessed using Q–Q plots and chi-square tests, followed by histograms with density functions of the adequate models and correlation coefficients. The findings indicate that the Singh Maddala Dagum distribution, provided the best fit in sixteen states, while the Weibull and Power-Pareto distributions each provided the best fit for only one state. The income inequality measures, including the Gini, Pietra, Bonferroni, Atkinson, generalized entropy, and Frigyes measures, were empirically calculated for each state and compared with theoretical estimates from the best-fitting model. The results revealed that Manipur exhibited the lowest values for these inequality measures, while Chhattisgarh showed the highest values, except for the empirical Frigyes measure <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\varphi \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>φ</mi> </math></EquationSource> </InlineEquation>, which peaked in Arunachal Pradesh.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparison of Income Inequality Among Indian States Using Quantile Functions

  • 期刊论文

  • 发表日期：

    2025年4月9日

查看全文
  • Ashlin Varkey,
  • Haritha N. Haridas

摘要

This study conducts a comparative analysis of quantile-based income distributions across Indian states, addressing a gap in the literature. It focuses on parametric models with closed-form quantile functions to model per capita household income data sourced from the India Human Development Survey II, excluding observations with negative or zero income and outliers. The Weibull, Power-Pareto, Singh-Maddala, Dagum, Singh Maddala Dagum, and Modified Lambda family were evaluated, with parameters estimated using the L-moments method. L-moments are advantageous because they are less affected by sampling variability and are robust to outliers. Model adequacy was first assessed using Q–Q plots and chi-square tests, followed by histograms with density functions of the adequate models and correlation coefficients. The findings indicate that the Singh Maddala Dagum distribution, provided the best fit in sixteen states, while the Weibull and Power-Pareto distributions each provided the best fit for only one state. The income inequality measures, including the Gini, Pietra, Bonferroni, Atkinson, generalized entropy, and Frigyes measures, were empirically calculated for each state and compared with theoretical estimates from the best-fitting model. The results revealed that Manipur exhibited the lowest values for these inequality measures, while Chhattisgarh showed the highest values, except for the empirical Frigyes measure \(\varphi \) φ , which peaked in Arunachal Pradesh.