arrow
返回

Propensity score analysis with missing data using a multi-task neural network

delete2023-02-15
delete2
delete
OA
AI
S
Shu Yang
P
Peipei Du
X
Xixi Feng
D
Daihai He
Y
Yaolong Chen
L
Linda L. D. Zhong
X
Xiaodong Yan *
J
Jiawei Luo *
DOI:10.1186/s12874-023-01847-2delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Background Propensity score analysis is increasingly used to control for confounding factors in observational studies. Unfortunately, unavoidable missing values make estimating propensity scores extremely challenging. We propose a new method for estimating propensity scores in data with missing values. Materials and methods Both simulated and real-world datasets are used in our experiments. The simulated datasets were constructed under 2 scenarios, the presence (T = 1) and the absence (T = 0) of the true effect. The real-world dataset comes from LaLonde's employment training program. We construct missing data with varying degrees of missing rates under three missing mechanisms: MAR, MCAR, and MNAR. Then we compare MTNN with 2 other traditional methods in different scenarios. The experiments in each scenario were repeated 20,000 times. Our code is publicly available at https://github.com/ljwa2323/MTNN. Results Under the three missing mechanisms of MAR, MCAR and MNAR, the RMSE between the effect and the true effect estimated by our proposed method is the smallest in simulations and in real-world data. Furthermore, the standard deviation of the effect estimated by our method is the smallest. In situations where the missing rate is low, the estimation of our method is more accurate. Conclusions MTNN can perform propensity score estimation and missing value filling at the same time through shared hidden layers and joint learning, which solves the dilemma of traditional methods and is very suitable for estimating true effects in samples with missing values. The method is expected to be broadly generalized and applied to real-world observational studies.
Keyword:
Observational study
Propensity score analysis
Neural network
Multitasking learning
Causal effect estimation
Inverse probability weighting
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

BMC Medical Research Methodology 封面图
BMC Medical Research Methodology
IF:
3.4
论文数:
4.0K
被引数:
3.5W

机构

南
南洋理工大学
学者数:
4.3K
论文数: 2.0K
被引数: 5.3W
成
成都医学院
学者数:
352
论文数: 109
被引数: 1.1K
兰州大学 封面图
兰州大学
学者数:
4.2K
论文数: 1.1K
被引数: 1.9W
香
香港理工大學
学者数:
2.1K
论文数: 1.0K
被引数: 4.1W
山东大学 封面图
山东大学
学者数:
8.1K
论文数: 2.2K
被引数: 4.2W
四川大学 封面图
四川大学
学者数:
1.2W
论文数: 2.7K
被引数: 5.6W
成都中医药大学 封面图
成都中医药大学
学者数:
1.4K
论文数: 338
被引数: 4.3K
学者 查看更多机构
引用论文

引用论文

Generating Synthetic Missing Data: A Review by Missing Mechanism
err2019-01-01
err55
errOAAI
errSantos, Miriam Seoane; Pereira, Ricardo Cardoso; Costa, Adriana Fonseca; Soares, Jastin Pompeu; Santos, Joao; Abreu, Pedro Henriques
err分享
err收藏
rTLS: Lightweight TLS Session Resumption for Constrained IoT Devices
err2020-11-28
err0
PREAI
errKoen Tange; David Howard; Travis Shanahan; Stefano Pepe; Xenofon Fafoutis; Nicola Dragoni
err分享
err收藏
Glioma cell VEGFR-2 confers resistance to chemotherapeutic and antiangiogenic treatments in PTEN-deficient glioblastoma
err2015-02-19
err0
errOAAI
errTobias Kessler; Felix Sahm; Jonas Blaes; Matthias Osswald; Petra Rübmann; David Milford; Severino Urban; Leonie Jestaedt; Sabine Heiland; Martin Bendszus; Anne Hertenstein; Philipp-Niclas Pfenning; Carmen Ruiz de Almodóvar; Antje Wick; Frank Winkler; Andreas von Deimling; Michael Platten; Wolfgang Wick; Markus Weiler
err分享
err收藏
Dissipation in quasistatically sheared wet and dry sand under confinement
err2012-08-24
err0
PREAI
errJ. E. Fiscina; M. Pakpour; A. Fall; N. Vandewalle; C. Wagner; D. Bonn
err分享
err收藏
Excision of a choledochal cyst and simultaneous hepatic lateral segmentectomy
err2013-09-21
err0
PREAI
errY. Tsuchida; F. Taniguchi; S. Nakahara; K. Uno; H. Kawarasaki; Y. Inoue; J. Nishikawa
err分享
err收藏
New 30-m Flexible Hybrid Energy Transfer Line With Liquid Hydrogen and Superconducting <inline-formula> <tex-math notation="TeX">$\hbox{MgB}_{2}$</tex-math></inline-formula> Cable&#x2014;Development and Test Results
err2015-06-01
err0
PREAI
errVitaly S. Vysotsky; Evegny V. Blagov; Valery V. Kostyuk; Alexander A. Nosov; Sergey S. Fetisov; Sergey Yu. Zanegin; Ilya V. Antyukhov; Valery P. Firsov; Grigory G. Svalov; Boris I. Katorgin; Vladimir S. Rachuk
err分享
err收藏
学者 查看更多内容