RMF2012 Abstracts
Session: Wednesday 4th July AM (09.15 - 12.45)
Title: Programme Evaluation Methods, and Inference in Difference-in-Difference Designs
Name: Mike Brewer
Affiliation: NCRM Node PEPA, Institute for Fiscal Studies
Abstract Details
In the first half, we review popular evaluation methods in policy analysis, showing the relationship between social experiments (randomised control trials), matching methods, instrumental variables, and control functions.
In the second half, we review the significant problems for inference (as captured in the estimated standard error) that can arise in the popular "difference-in-difference" designs. These arise especially where data have a multi-level structure, and there is serial correlation in the treatment or policy variation, and in the group-level shocks. We discuss best-practice in dealing with such problems.
Presentation downloads
Presenter: Mike Brewer
Getting inference right when using difference-in-differences
Presenter: Monica Costa dias
The level of the session is: Accessible
Presentation details
Presenter 1
Start time: 09:15
Presentation title: Programme Evaluation Methods
Author: Dr Monica Costa Dias
Affiliation: Institute for Fiscal Studies
Presenter 2
Start time: 9.15
Presentation title: Programme Evaluation Methods
Author: Dr Adam Rosen
Affiliation: UCL and Institute for Fiscal Studies
Presenter 3
Start time: 09:15
Presentation title: Getting inference right in "difference-in-difference" designs
Author: Professor Mike Brewer
Affiliation: University of Essex and Institute for Fiscal Studies
Presenter 4
Start time: 11:15
Presentation title: Getting inference right in "difference-in-difference" designs
Author: Mr Robert Joyce
Affiliation: Institute for Fiscal Studies