Abstract
By treating one nuisance parameter as two distinct parameters and using some moment conditions twice, we provide an explanation to a paradox in the inverse probability weighting estimation of the average treatment effect on the treated.
Original language | English |
---|---|
Pages (from-to) | 163-165 |
Number of pages | 3 |
Journal | Economics Letters |
Volume | 110 |
Issue number | 2 |
DOIs | |
Publication status | Published - 2011 Feb |
Keywords
- Average treatment effect
- Average treatment effect on the treated
- Estimated propensity score
- Generalized method of moments
- Method of moments
- Paradox
ASJC Scopus subject areas
- Finance
- Economics and Econometrics