Estimation of the Conditional Variance in Paired Experiments

By Alberto Abadie, Guido W. Imbens. 2008

69 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/0423c2bb06b5be74c403b60265fe9712881a29d9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

In Statistics and Probability

Citations in the corpus

Optimal bandwidth choice for robust bias-corrected inference in regression discontinuity designs (2018, methodology): “A second option is to use a nearest-neighbor-based variance estimators with a fixed number of neighbors, following the ideas of Muller and Stadtmuller (1987) and Abadie and Imbens (2008). To define these, let J be a fixed number and j(i) be the j-th closest observation to Xi, j = 1, . . . , J , and ”
Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference (2007, methodology): “…forms of matching combined with semiparametric (kernel weighting) analyses, as well as to the parametric bias adjustment for one form of matching by Abadie and Imbens (2006b), except that, to avoid inducing new biases, we recommend below that matching be evaluated prior to examining the dependent…”
Generalized Optimal Matching Methods for Causal Inference (2016, methodology): “[42] Alberto Abadie and Guido W Imbens.”
Regression-adjusted average treatment effect estimates in stratified randomized experiments (2019, background): “Suitable variance estimators can be found in Abadie and Imbens [1], Imai [15], and Fogarty [9, 10].”
On the Effect of Bias Estimation on Coverage Accuracy in Nonparametric Inference (2015, methodology): “Another option is to use a nearest-neighbor-based variance estimators with a fixed number of neighbors, following the ideas of Muller and Stadtmuller (1987) and Abadie and Imbens (2008). Note that none of these estimators assume local or global homoskedasticity nor rely on new tuning parameters.”
Adjusting treatment effect estimates by post‐stratification in randomized experiments (2013, background): “Sometimes there can be nice tricks—Alberto and IMBENS (2008), for example, estimated these parameters for matched-pairs by looking at pairs of pairs matched on covariates—but generally bounding the standard error is the best one can do.”
Rdrobust: Software for Regression-discontinuity Designs (2017, methodology): “Nearest-neighbor method This method uses ideas in M¨uller and Stadtmuller (1987) and Abadie and Imbens (2008).”
Optimal a priori balance in the design of controlled experiments (2013, background): “or any pair-matched allocation. But, in such designs, other estimators of the error are in fact appropriate, enabling an analysis tailored for the design ( Fisher, 1935 ; Snedecor and Cochran, 1989 ; Abadie and Imbens, 2008 ). (Note that, in light of Thm. 2, variance of e ! ect estimation must be ex”