Double Machine Learning based Program Evaluation under Unconfoundedness

This paper reviews, applies and extends recently proposed methods based on Double Machine Learning (DML) with a focus on program evaluation under unconfoundedness. DML based methods leverage flexible prediction models to adjust for confounding variables in the estimation of (i) standard average effe...

Full description

Bibliographic Details
Main Author: Knaus, Michael C.
Format: Article in Journal/Newspaper
Language:unknown
Published: arXiv 2020
Subjects:
Econometrics econ.EM
FOS Economics and business
DML
Online Access:https://dx.doi.org/10.48550/arxiv.2003.03191
https://arxiv.org/abs/2003.03191
id ftdatacite:10.48550/arxiv.2003.03191
record_format openpolar
spelling ftdatacite:10.48550/arxiv.2003.03191 2023-05-15T16:01:17+02:00 Double Machine Learning based Program Evaluation under Unconfoundedness Knaus, Michael C. 2020 https://dx.doi.org/10.48550/arxiv.2003.03191 https://arxiv.org/abs/2003.03191 unknown arXiv arXiv.org perpetual, non-exclusive license http://arxiv.org/licenses/nonexclusive-distrib/1.0/ Econometrics econ.EM FOS Economics and business Article CreativeWork article Preprint 2020 ftdatacite https://doi.org/10.48550/arxiv.2003.03191 2022-03-10T16:03:47Z This paper reviews, applies and extends recently proposed methods based on Double Machine Learning (DML) with a focus on program evaluation under unconfoundedness. DML based methods leverage flexible prediction models to adjust for confounding variables in the estimation of (i) standard average effects, (ii) different forms of heterogeneous effects, and (iii) optimal treatment assignment rules. An evaluation of multiple programs of the Swiss Active Labour Market Policy illustrates how DML based methods enable a comprehensive program evaluation. Motivated by extreme individualised treatment effect estimates of the DR-learner, we propose the normalised DR-learner (NDR-learner) to address this issue. The NDR-learner acknowledges that individualised effect estimates can be stabilised by an individualised normalisation of inverse probability weights. Article in Journal/Newspaper DML DataCite Metadata Store (German National Library of Science and Technology)
institution Open Polar
collection DataCite Metadata Store (German National Library of Science and Technology)
op_collection_id ftdatacite
language unknown
topic Econometrics econ.EM
FOS Economics and business
spellingShingle Econometrics econ.EM
FOS Economics and business
Knaus, Michael C.
Double Machine Learning based Program Evaluation under Unconfoundedness
topic_facet Econometrics econ.EM
FOS Economics and business
description This paper reviews, applies and extends recently proposed methods based on Double Machine Learning (DML) with a focus on program evaluation under unconfoundedness. DML based methods leverage flexible prediction models to adjust for confounding variables in the estimation of (i) standard average effects, (ii) different forms of heterogeneous effects, and (iii) optimal treatment assignment rules. An evaluation of multiple programs of the Swiss Active Labour Market Policy illustrates how DML based methods enable a comprehensive program evaluation. Motivated by extreme individualised treatment effect estimates of the DR-learner, we propose the normalised DR-learner (NDR-learner) to address this issue. The NDR-learner acknowledges that individualised effect estimates can be stabilised by an individualised normalisation of inverse probability weights.
format Article in Journal/Newspaper
author Knaus, Michael C.
author_facet Knaus, Michael C.
author_sort Knaus, Michael C.
title Double Machine Learning based Program Evaluation under Unconfoundedness
title_short Double Machine Learning based Program Evaluation under Unconfoundedness
title_full Double Machine Learning based Program Evaluation under Unconfoundedness
title_fullStr Double Machine Learning based Program Evaluation under Unconfoundedness
title_full_unstemmed Double Machine Learning based Program Evaluation under Unconfoundedness
title_sort double machine learning based program evaluation under unconfoundedness
publisher arXiv
publishDate 2020
url https://dx.doi.org/10.48550/arxiv.2003.03191
https://arxiv.org/abs/2003.03191
genre DML
genre_facet DML
op_rights arXiv.org perpetual, non-exclusive license
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
op_doi https://doi.org/10.48550/arxiv.2003.03191
_version_ 1766397215834636288

Similar Items