No Detail Found!

Quantile Regression Analysis of Monthly Earnings in Pakistan.

Author(s)
I. A. ARSHAD Department of Statistics, Allama Iqbal Open University,Islamabad, Pakistan
U. YOUNAS Department of Statistics, Allama Iqbal Open University,Islamabad, Pakistan
A. W. SHAIKH Institute of Mathematics and Computer Science, University of Sindh, Jamshoroo, Pakistan
M. S. CHANDIO Institute of Mathematics and Computer Science, University of Sindh, Jamshoroo, Pakistan
Abstract
In this study, we empirically analyze the monthly earning distribution of Pakistan. The log of monthly earning is taken as a response variable, while education, experience, age, sex, marital status, nature of work, region, and the provinces are used as explanatory variables. Ordinary least square regression and quantile regression techniques are used to estimate the relationship among these variables. Quantile regression, instead of the point estimate of the conditional mean, can be used to estimate the whole distribution, especially the upper tail and lower tail which we are interested in. The comparison of OLS, and quantile regression shows that quantile regression can provide more informative estimation results. We also use quantile regression's equivariant property to transform our response variable from log to level.
Publication Details
Page(s) 919-924
DOI DOI not available
Published Journal: Sindh University Research Journal, Volume: 48, Issue: 4, Year: 2016

Request the full text from PASTIC

Keywords
OLS regression, earnings distribution
0

Citations

0

Downloads

35

Views