Advanced

Income inequality regression models with applications

Mehic, Adrian LU (2017) STAM01 20162
Department of Statistics
Abstract
This thesis addresses three topics in income inequality. First, a cross-
sectional dataset of 30 countries is used to investigate the causes of between-
country differences in income inequality. Secondly, a panel of the same
countries is used to examine the drivers behind the change in inequality
between 1985 and 2013. This part of the thesis utilizes dynamic regres-
sion models for panel data, and compares the estimates from the two most
common dynamic panel models. The problem of endogenity of explana-
tory variables is addressed, and possible solutions to this problem are dis-
cussed, with an emphasis on techniques based on the generalized method
of moments (GMM). The main findings are that the trade-to-GDP ratio,
the... (More)
This thesis addresses three topics in income inequality. First, a cross-
sectional dataset of 30 countries is used to investigate the causes of between-
country differences in income inequality. Secondly, a panel of the same
countries is used to examine the drivers behind the change in inequality
between 1985 and 2013. This part of the thesis utilizes dynamic regres-
sion models for panel data, and compares the estimates from the two most
common dynamic panel models. The problem of endogenity of explana-
tory variables is addressed, and possible solutions to this problem are dis-
cussed, with an emphasis on techniques based on the generalized method
of moments (GMM). The main findings are that the trade-to-GDP ratio,
the industrial employment share and the political color of the government
are the most important explanatory variables of income inequality over
time. The final question addressed in the thesis concerns cross-country
income inequality convergence over time. Notably, inequality is shown to
converge at slightly slower rate than reported by previous studies. (Less)
Please use this url to cite or link to this publication:
author
Mehic, Adrian LU
supervisor
organization
course
STAM01 20162
year
type
H1 - Master's Degree (One Year)
subject
keywords
dynamic panel data, income inequality, endogenity, Gini coefficient
language
English
id
8904917
date added to LUP
2017-05-04 09:24:27
date last changed
2017-05-04 09:24:27
@misc{8904917,
  abstract     = {This thesis addresses three topics in income inequality. First, a cross-
sectional dataset of 30 countries is used to investigate the causes of between-
country differences in income inequality. Secondly, a panel of the same
countries is used to examine the drivers behind the change in inequality
between 1985 and 2013. This part of the thesis utilizes dynamic regres-
sion models for panel data, and compares the estimates from the two most
common dynamic panel models. The problem of endogenity of explana-
tory variables is addressed, and possible solutions to this problem are dis-
cussed, with an emphasis on techniques based on the generalized method
of moments (GMM). The main findings are that the trade-to-GDP ratio,
the industrial employment share and the political color of the government
are the most important explanatory variables of income inequality over
time. The final question addressed in the thesis concerns cross-country
income inequality convergence over time. Notably, inequality is shown to
converge at slightly slower rate than reported by previous studies.},
  author       = {Mehic, Adrian},
  keyword      = {dynamic panel data,income inequality,endogenity,Gini coefficient},
  language     = {eng},
  note         = {Student Paper},
  title        = {Income inequality regression models with applications},
  year         = {2017},
}