Nonparametric Estimation of Average Treatment Effects under Exogeneity: A Review
read more
Citations
An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies
Some practical guidance for the implementation of propensity score matching
Matching Methods for Causal Inference: A Review and a Look Forward
Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference
Recent developments in the econometrics of program evaluation
References
An introduction to the bootstrap
The central role of the propensity score in observational studies for causal effects
Experimental and Quasi-Experimental Designs for Generalized Causal Inference
Estimating causal effects of treatments in randomized and nonrandomized studies.
Constructing a Control Group Using Multivariate Matched Sampling Methods That Incorporate the Propensity Score
Related Papers (5)
The central role of the propensity score in observational studies for causal effects
Matching As An Econometric Evaluation Estimator: Evidence from Evaluating a Job Training Programme
Frequently Asked Questions (8)
Q2. What is the key feature that casts doubt on the relevance of the asymptotic?
A key feature that casts doubt on the relevance of the asymptotic distributions is that the =N consistency is obtained by averaging a nonparametric estimator of a regression function that itself has a slow nonparametric convergence rate over the empirical distribution of its argument.
Q3. What methods were used to create nonexperimental estimates of the average treatment effect?
He then used a number of methods—ranging from a simple difference, to least squares adjustment, a Heckman selection correction, and differenceindifferences techniques—to create nonexperimental estimates of the average treatment effect.
Q4. How can one undersmooth the estimation of the propensity score?
In practice one may wish to undersmooth the estimation of the propensity score, either by choosing a bandwidth smaller than optimal for nonparametric estimation or by including higher-order terms in a series expansion.
Q5. What is the third method to estimate the regression function for the controls?
The third method is to estimate the same regression function for the controls, but using only those that are used as matches for the treated units, with weights corresponding to the number of times a control observations is used as a match (see Abadie and Imbens, 2002).
Q6. What is the way to estimate the regression function m1( x)?
[If unit i is a control unit, the correction will be done using an estimator for the regression function m1( x) based on a linear speci cation Y i 5 a1 1 b91X i estimated on the treated units.]
Q7. What is the relative merits of these estimators?
In practice the relative merits of these estimators will depend on whether the propensity score is more or less smooth than the regression functions, and on whether additional information is available about either the propensity score or the regression functions.
Q8. What is the second assumption regarding the conditional regression functions and the propensity score?
the authors make a second assumption regarding the joint distribution of treatments and covariates:ASSUMPTION 2.2 (OVERLAP):0 , Pr~W 5 1uX! , 1.For many of the formal results one will also need smoothness assumptions on the conditional regression functions and the propensity score [mw( x) and e( x)], and moment conditions on Y(w).