BAYESIAN STRUCTURED ADDITIVE DISTRIBUTIONAL REGRESSION WITH AN APPLICATION TO REGIONAL INCOME INEQUALITY IN GERMANY

被引:87
作者
Klein, Nadja [1 ]
Kneib, Thomas [1 ]
Lang, Stefan [2 ]
Sohn, Alexander [1 ]
机构
[1] Univ Gottingen, Chair Stat, D-37073 Gottingen, Germany
[2] Univ Innsbruck, Dept Stat, A-6020 Innsbruck, Austria
关键词
Generalised additive models for location; scale and shape; income distribution; iteratively weighted least squares proposal; Markov chain Monte Carlo; semiparametric regression; wage inequality; MODEL; DECOMPOSITION; SELECTION; SPECIFICATION; POLARIZATION; LOCATION; SCALE;
D O I
10.1214/15-AOAS823
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
We propose a generic Bayesian framework for inference in distributional regression models in which each parameter of a potentially complex response distribution and not only the mean is related to a structured additive predictor. The latter is composed additively of a variety of different functional effect types such as nonlinear effects, spatial effects, random coefficients, interaction surfaces or other (possibly nonstandard) basis function representations. To enforce specific properties of the functional effects such as smoothness, informative multivariate Gaussian priors are assigned to the basis function coefficients. Inference can then be based on computationally efficient Markov chain Monte Carlo simulation techniques where a generic proceduremakes use of distribution-specific iteratively weighted least squares approximations to the full conditionals. The framework of distributional regression encompasses many special cases relevant for treating nonstandard response structures such as highly skewed nonnegative responses, overdispersed and zero-inflated counts or shares including the possibility for zero-and one-inflation. We discuss distributional regression along a study on determinants of labour incomes for full-time working males in Germany with a particular focus on regional differences after the German reunification. Controlling for age, education, work experience and local disparities, we estimate full conditional income distributions allowing us to study various distributional quantities such as moments, quantiles or inequality measures in a consistent manner in one joint model. Detailed guidance on practical aspects of model choice including the selection of several competing distributions for labour incomes and the consideration of different covariate effects on the income distribution complete the distributional regression analysis. We find that next to a lower expected income, full-time working men in East Germany also face a more unequal income distribution than men in the West, ceteris paribus.
引用
收藏
页码:1024 / 1052
页数:29
相关论文
共 70 条
[1]  
[Anonymous], HDB INCOME INEQUALIT
[2]  
[Anonymous], 5708 IZA
[3]  
[Anonymous], BAMLSS BAYESIAN ADDI
[4]  
[Anonymous], VERBR DEUTSCHL LANG
[5]  
[Anonymous], 2005, MONOGRAPHS STAT APPL, DOI DOI 10.1201/9780203492024
[6]  
[Anonymous], BAYESIAN STRUCTURE S
[7]  
[Anonymous], J R STAT C IN PRESS
[8]  
[Anonymous], BAYESIAN STRUCTURE S
[9]  
Arnold BC, 2008, ECON STUD INEQUAL SO, V5, P119, DOI 10.1007/978-0-387-72796-7_7
[10]   FROM BOTTOM TO TOP: THE ENTIRE INCOME DISTRIBUTION IN GERMANY, 1992-2003 [J].
Bach, Stefan ;
Corneo, Giacomo ;
Steiner, Viktor .
REVIEW OF INCOME AND WEALTH, 2009, 55 (02) :303-330