Interpretable Almost-Matching-Exactly With Instrumental Variables

被引:0
|
作者
Awan, M. Usaid [1 ]
Liu, Yameng [1 ]
Morucci, Marco [1 ]
Roy, Sudeepa [1 ]
Rudin, Cynthia [1 ]
Volfovsky, Alexander [1 ]
机构
[1] Duke Univ, Durham, NC 27708 USA
关键词
IDENTIFICATION; INFERENCE; MALARIA;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Uncertainty in the estimation of the causal effect in observational studies is often due to unmeasured confounding, i.e., the presence of unobserved covariates linking treatments and outcomes. Instrumental Variables (IV) are commonly used to reduce the effects of unmeasured confounding. Existing methods for IV estimation either require strong parametric assumptions, use arbitrary distance metrics, or do not scale well to large datasets. We propose a matching framework for IV in the presence of observed categorical confounders that addresses these weaknesses. Our method first matches units exactly, and then consecutively drops variables to approximately match the remaining units on as many variables as possible. We show that our algorithm constructs better matches than other existing methods on simulated datasets, and we produce interesting results in an application to political canvassing.
引用
收藏
页码:1116 / 1126
页数:11
相关论文
共 50 条
  • [1] Almost-Matching-Exactly for Treatment Effect Estimation under Network Interference
    Awan, M. Usaid
    Roy, Sudeepa
    Morucci, Marco
    Rudin, Cynthia
    Orlandi, Vittorio
    Volfovsky, Alexander
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 108, 2020, 108 : 3252 - 3261
  • [2] Should instrumental variables be used as matching variables?
    Wooldridge, Jeffrey M.
    RESEARCH IN ECONOMICS, 2016, 70 (02) : 232 - 237
  • [3] Interpretable Almost-Exact Matching for Causal Inference
    Dieng, Awa
    Liu, Yameng
    Roy, Sudeepa
    Rudin, Cynthia
    Volfovsky, Alexander
    22ND INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 89, 2019, 89
  • [4] AME: Interpretable Almost Exact Matching for Causal Inference
    Jiang, Haoning
    Howell, Thomas
    Gupta, Neha R.
    Orlandi, Vittorio
    Morucci, Marco
    Parikh, Harsh
    Roy, Sudeepa
    Rudin, Cynthia
    Volfovsky, Alexander
    NEURIPS 2021 COMPETITIONS AND DEMONSTRATIONS TRACK, VOL 176, 2021, 176 : 355 - 359
  • [5] Propensity-score matching with instrumental variables
    Ichimura, H
    Taber, C
    AMERICAN ECONOMIC REVIEW, 2001, 91 (02): : 119 - 124
  • [6] The distribution of the sum of independent random variables can be calculated exactly almost always
    Oldenbuerger, Hartmut A.
    INTERNATIONAL JOURNAL OF PSYCHOLOGY, 2008, 43 (3-4) : 647 - 647
  • [7] Near/far matching: a study design approach to instrumental variables
    Baiocchi, Mike
    Small, Dylan S.
    Yang, Lin
    Polsky, Daniel
    Groeneveld, Peter W.
    HEALTH SERVICES AND OUTCOMES RESEARCH METHODOLOGY, 2012, 12 (04) : 237 - 253
  • [8] Near/far matching: a study design approach to instrumental variables
    Mike Baiocchi
    Dylan S. Small
    Lin Yang
    Daniel Polsky
    Peter W. Groeneveld
    Health Services and Outcomes Research Methodology, 2012, 12 (4) : 237 - 253
  • [9] Exactly distribution-free inference in instrumental variables regression with possibly weak instruments
    Andrews, Donald W. K.
    Marmer, Vadim
    JOURNAL OF ECONOMETRICS, 2008, 142 (01) : 183 - 200
  • [10] FLAME: A fast large-scale almost matching exactly approach to causal inference
    Wang, Tianyu
    Morucci, Marco
    Awan, M. Usaid
    Liu, Yameng
    Roy, Sudeepa
    Rudin, Cynthia
    Volfovsky, Alexander
    Journal of Machine Learning Research, 2021, 22