Adaptive Experimental Design Using the Propensity Score

Many social experiments are run in multiple waves or replicate earlier social experiments. In principle, the sampling design can be modified in later stages or replications to allow for more efficient estimation of causal effects. We consider the design of a two-stage experiment for estimating an av...

Full description

Saved in:
Bibliographic Details
Published in:Journal of business & economic statistics Vol. 29; no. 1; pp. 96 - 108
Main Authors: Hahn, Jinyong, Hirano, Keisuke, Karlan, Dean
Format: Journal Article
Language:English
Published: Alexandria Taylor & Francis 01.01.2011
American Statistical Association
Taylor & Francis Ltd
Subjects:
Asymptotic methods
Conditional probabilities
Design efficiency
Design of experiments
Economic statistics
Efficiency bound
Estimating techniques
Estimators
Experiment design
Experimental design
Fundraising
Hunger
Probability
Propensity score
Sample size
Software
Statistical discrepancies
Statistical variance
Studies
ISSN:0735-0015, 1537-2707
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Many social experiments are run in multiple waves or replicate earlier social experiments. In principle, the sampling design can be modified in later stages or replications to allow for more efficient estimation of causal effects. We consider the design of a two-stage experiment for estimating an average treatment effect when covariate information is available for experimental subjects. We use data from the first stage to choose a conditional treatment assignment rule for units in the second stage of the experiment. This amounts to choosing the propensity score, the conditional probability of treatment given covariates. We propose to select the propensity score to minimize the asymptotic variance bound for estimating the average treatment effect. Our procedure can be implemented simply using standard statistical software and has attractive large-sample properties.
Bibliography:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:0735-0015
1537-2707
DOI:10.1198/jbes.2009.08161