2020/09/08 by Maike Hohberg, Hohberg, Maike, Francesco Donat +5
Economics, Econometrics and Finance · Social Sciences · #Applications (stat.AP) #Economic and Environmental Valuation #FOS: Computer and information sciences #Income, Poverty, and Inequality #Methodology (stat.ME) #Spatial and Panel Data Analysis
paper · pdf · doi:10.48550/arxiv.2009.03646
openalex publication_date 2020/09/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Poverty is a multidimensional concept often comprising a monetary outcome and\nother welfare dimensions such as education, subjective well-being or health,\nthat are measured on an ordinal scale. In applied research, multidimensional\npoverty is ubiquitously assessed by studying each poverty dimension\nindependently in univariate regression models or by combining several poverty\ndimensions into a scalar index. This inhibits a thorough analysis of the\npotentially varying interdependence between the poverty dimensions. We propose\na multivariate copula generalized additive model for location, scale and shape\n(copula GAMLSS or distributional copula model) to tackle this challenge. By\nrelating the copula parameter to covariates, we specifically examine if certain\nfactors determine the dependence between poverty dimensions. Furthermore,\nspecifying the full conditional bivariate distribution, allows us to derive\nseveral features such as poverty risks and dependence measures coherently from\none model for different individuals. We demonstrate the approach by studying\ntwo important poverty dimensions: income and education. Since the level of\neducation is measured on an ordinal scale while income is continuous, we extend\nthe bivariate copula GAMLSS to the case of mixed ordered-continuous outcomes.\nThe new model is integrated into the GJRM package in R and applied to data from\nIndonesia. Particular emphasis is given to the spatial variation of the\nincome-education dependence and groups of individuals at risk of being\nsimultaneously poor in both education and income dimensions.\n