This paper proposes a sensitivity analysis test of unobservable selection for matching estimators based on a "leave-one-covariate-out" (LOCO) algorithm. Rooted in the machine learning literature, this sensitivity test performs a bootstrap over different subsets of covariates, and simulates various estimation scenarios to be compared with the baseline matching results. We provide an empirical application, comparing results with more traditional sensitivity tests. (C) 2019 Published by Elsevier B.V.
Data-driven sensitivity analysis for matching estimators
Cerulli;Giovanni
2019
Abstract
This paper proposes a sensitivity analysis test of unobservable selection for matching estimators based on a "leave-one-covariate-out" (LOCO) algorithm. Rooted in the machine learning literature, this sensitivity test performs a bootstrap over different subsets of covariates, and simulates various estimation scenarios to be compared with the baseline matching results. We provide an empirical application, comparing results with more traditional sensitivity tests. (C) 2019 Published by Elsevier B.V.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.