Causal Matching using Random Hyperplane Tessellations

Abhishek Dalvi, Neil Ashtekar, Vasant G Honavar
Proceedings of the Third Conference on Causal Learning and Reasoning, PMLR 236:688-702, 2024.

Abstract

Matching is one of the simplest approaches for estimating causal effects from observational data. Matching techniques compare the observed outcomes across pairs of individuals with similar covariate values but different treatment statuses in order to estimate causal effects. However, traditional matching techniques are unreliable given high-dimensional covariates due to the infamous curse of dimensionality. To overcome this challenge, we propose a simple, fast, yet highly effective approach to matching using Random Hyperplane Tessellations (RHPT). First, we prove that the RHPT representation is an approximate balancing score – thus maintaining the strong ignorability assumption – and provide empirical evidence for this claim. Second, we report results of extensive experiments showing that matching using RHPT outperforms traditional matching techniques and is competitive with state-of-the-art deep learning methods for causal effect estimation. In addition, RHPT avoids the need for computationally expensive training of deep neural networks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v236-dalvi24a, title = {Causal Matching using Random Hyperplane Tessellations}, author = {Dalvi, Abhishek and Ashtekar, Neil and Honavar, Vasant G}, booktitle = {Proceedings of the Third Conference on Causal Learning and Reasoning}, pages = {688--702}, year = {2024}, editor = {Locatello, Francesco and Didelez, Vanessa}, volume = {236}, series = {Proceedings of Machine Learning Research}, month = {01--03 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v236/dalvi24a/dalvi24a.pdf}, url = {https://proceedings.mlr.press/v236/dalvi24a.html}, abstract = {Matching is one of the simplest approaches for estimating causal effects from observational data. Matching techniques compare the observed outcomes across pairs of individuals with similar covariate values but different treatment statuses in order to estimate causal effects. However, traditional matching techniques are unreliable given high-dimensional covariates due to the infamous curse of dimensionality. To overcome this challenge, we propose a simple, fast, yet highly effective approach to matching using Random Hyperplane Tessellations (RHPT). First, we prove that the RHPT representation is an approximate balancing score – thus maintaining the strong ignorability assumption – and provide empirical evidence for this claim. Second, we report results of extensive experiments showing that matching using RHPT outperforms traditional matching techniques and is competitive with state-of-the-art deep learning methods for causal effect estimation. In addition, RHPT avoids the need for computationally expensive training of deep neural networks.} }
Endnote
%0 Conference Paper %T Causal Matching using Random Hyperplane Tessellations %A Abhishek Dalvi %A Neil Ashtekar %A Vasant G Honavar %B Proceedings of the Third Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2024 %E Francesco Locatello %E Vanessa Didelez %F pmlr-v236-dalvi24a %I PMLR %P 688--702 %U https://proceedings.mlr.press/v236/dalvi24a.html %V 236 %X Matching is one of the simplest approaches for estimating causal effects from observational data. Matching techniques compare the observed outcomes across pairs of individuals with similar covariate values but different treatment statuses in order to estimate causal effects. However, traditional matching techniques are unreliable given high-dimensional covariates due to the infamous curse of dimensionality. To overcome this challenge, we propose a simple, fast, yet highly effective approach to matching using Random Hyperplane Tessellations (RHPT). First, we prove that the RHPT representation is an approximate balancing score – thus maintaining the strong ignorability assumption – and provide empirical evidence for this claim. Second, we report results of extensive experiments showing that matching using RHPT outperforms traditional matching techniques and is competitive with state-of-the-art deep learning methods for causal effect estimation. In addition, RHPT avoids the need for computationally expensive training of deep neural networks.
APA
Dalvi, A., Ashtekar, N. & Honavar, V.G.. (2024). Causal Matching using Random Hyperplane Tessellations. Proceedings of the Third Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 236:688-702 Available from https://proceedings.mlr.press/v236/dalvi24a.html.

Related Material