A Stable and Efficient Covariate-Balancing Estimator for Causal Survival Effects

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pham, Khiem, Hirshberg, David A., Huynh-Pham, Phuong-Mai, Santacatterina, Michele, Lim, Ser-Nam, Zabih, Ramin
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909204716978176
author Pham, Khiem
Hirshberg, David A.
Huynh-Pham, Phuong-Mai
Santacatterina, Michele
Lim, Ser-Nam
Zabih, Ramin
author_facet Pham, Khiem
Hirshberg, David A.
Huynh-Pham, Phuong-Mai
Santacatterina, Michele
Lim, Ser-Nam
Zabih, Ramin
contents We propose an empirically stable and asymptotically efficient covariate-balancing approach to the problem of estimating survival causal effects in data with conditionally-independent censoring. This addresses a challenge often encountered in state-of-the-art nonparametric methods: the use of inverses of small estimated probabilities and the resulting amplification of estimation error. We validate our theoretical results in experiments on synthetic and semi-synthetic data.
format Preprint
id arxiv_https___arxiv_org_abs_2310_02278
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Stable and Efficient Covariate-Balancing Estimator for Causal Survival Effects
Pham, Khiem
Hirshberg, David A.
Huynh-Pham, Phuong-Mai
Santacatterina, Michele
Lim, Ser-Nam
Zabih, Ramin
Methodology
Machine Learning
We propose an empirically stable and asymptotically efficient covariate-balancing approach to the problem of estimating survival causal effects in data with conditionally-independent censoring. This addresses a challenge often encountered in state-of-the-art nonparametric methods: the use of inverses of small estimated probabilities and the resulting amplification of estimation error. We validate our theoretical results in experiments on synthetic and semi-synthetic data.
title A Stable and Efficient Covariate-Balancing Estimator for Causal Survival Effects
topic Methodology
Machine Learning
url https://arxiv.org/abs/2310.02278