Robust Bayesian causal estimation for causal inference in medical diagnosis

被引:0
|
作者
Basu, Tathagata [1 ]
Troffaes, Matthias C. M. [2 ]
机构
[1] Univ Strathclyde, Civil & Environm Engn, 16 Richmond St, Glasgow G1 1XQ, Scotland
[2] Univ Durham, Dept Math Sci, South Rd, Durham DH1 3LE, England
关键词
High dimensional regression; Variable selection; Bayesian analysis; Imprecise probability; VARIABLE SELECTION; PROPENSITY SCORE; UNCERTAINTY; LASSO;
D O I
10.1016/j.ijar.2024.109330
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Causal effect estimation is a critical task in statistical learning that aims to find the causal effect on subjects by identifying causal links between a number of predictor (or, explanatory) variables and the outcome of a treatment. In a regressional framework, we assign a treatment and outcome model to estimate the average causal effect. Additionally, for high dimensional regression problems, variable selection methods are also used to find a subset of predictor variables that maximises the predictive performance of the underlying model for better estimation of the causal effect. In this paper, we propose a different approach. We focus on the variable selection aspects of high dimensional causal estimation problem. We suggest a cautious Bayesian group LASSO (least absolute shrinkage and selection operator) framework for variable selection using prior sensitivity analysis. We argue that in some cases, abstaining from selecting (or, rejecting) a predictor is beneficial and we should gather more information to obtain a more decisive result. We also show that for problems with very limited information, expert elicited variable selection can give us a more stable causal effect estimation as it avoids overfitting. Lastly, we carry a comparative study with synthetic dataset and show the applicability of our method in real-life situations.
引用
收藏
页数:16
相关论文
共 50 条
  • [1] A Bayesian view of doubly robust causal inference
    Saarela, O.
    Belzile, L. R.
    Stephens, D. A.
    BIOMETRIKA, 2016, 103 (03) : 667 - 681
  • [2] Relaxed doubly robust estimation in causal inference
    Xu, Tinghui
    Zhao, Jiwei
    STATISTICAL THEORY AND RELATED FIELDS, 2024, 8 (01) : 69 - 79
  • [3] Robust variance estimation and inference for causal effect estimation
    Tran, Linh
    Petersen, Maya
    Schwab, Joshua
    van der Laan, Mark J.
    JOURNAL OF CAUSAL INFERENCE, 2023, 11 (01)
  • [4] Multiply robust Bayesian procedures for causal inference problems
    Gochanour, Benjamin
    Chen, Sixia
    Beebe, Laura
    COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2024, 53 (07) : 3271 - 3284
  • [5] Semiparametric Bayesian doubly robust causal estimation
    Luo, Yu
    Graham, Daniel J.
    McCoy, Emma J.
    JOURNAL OF STATISTICAL PLANNING AND INFERENCE, 2023, 225 : 171 - 187
  • [6] Doubly robust estimation of the causal effects in the causal inference with missing outcome data
    Han F.
    Journal of Ambient Intelligence and Humanized Computing, 2024, 15 (Suppl 1) : 11 - 11
  • [7] An Alternative Doubly Robust Estimation in Causal Inference Model
    Wei, Shaojie
    Li, Gaorong
    Zhang, Zhongzhan
    COMMUNICATIONS IN MATHEMATICS AND STATISTICS, 2024, 12 (04) : 659 - 678
  • [8] Active Bayesian Causal Inference
    Toth, Christian
    Lorch, Lars
    Knoll, Christian
    Krause, Andreas
    Pernkopf, Franz
    Peharz, Robert
    von Kuegelgen, Julius
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35 (NEURIPS 2022), 2022,
  • [9] SEMIPARAMETRIC BAYESIAN CAUSAL INFERENCE
    Ray, Kolyan
    van der Vaart, Aad
    ANNALS OF STATISTICS, 2020, 48 (05): : 2999 - 3020
  • [10] Doubly robust estimation in missing data and causal inference models
    Bang, H
    BIOMETRICS, 2005, 61 (04) : 962 - 972