Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.773484
Title: Three applications using panel data
Author: Houston, John Alexander
ISNI:       0000 0004 7960 8691
Awarding Body: University of Strathclyde
Current Institution: University of Strathclyde
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
This Thesis comprises three applied studies related to labour economics that utilise different Panel Datasets, assembled by the Author from public sources. The common Aim in all three studies is to determine whether the conclusions drawn from regression-based modelling would have been significantly different had standard estimators, as opposed to Panel-based ones, been applied. The stance taken, is that of the policymaker wishing to either evaluate the effectiveness of an established policy, or to quantify an established trend in Society, with a view to designing and implementing Policy in the future. In so doing, they wish to adopt an evidence-based approach to this exercise, using historic data. While the three studies are related to labour economics, the have been deliberately chosen to differ from each other, both in the scenario being studied, the variables involved, and the estimators applied. The first study is related to gendered policing and arrests made for gender-based violence in England and Wales; the second, estimates the returns to qualifications in the UK Labour market, evaluated at the quantiles of males' and females' wage-rate distributions and the third, the technical efficiency of a panel of US Airlines. The overall conclusions from the studies are that Panel-based estimators can lead to significantly different conclusions being drawn from regression-based evaluations but this need not always be the case. Nevertheless, where data are available as a Panel, or can be rendered as such, the Policy-maker should apply both Pooled- and Panel based estimators before drawing their conclusions from the exercise.
Supervisor: Wright, Robert E. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.773484  DOI:
Share: