Higher-Order Least Squares: Assessing Partial Goodness of Fit of Linear Causal Models

被引：5

作者：

Schultheiss, Christoph ^{[1
]}

Buhlmann, Peter ^{[1
]}

Yuan, Ming ^{[2
]}

机构：

[1] Swiss Fed Inst Technol, Seminar Stat, Zurich, Switzerland

[2] Columbia Univ, Dept Stat, New York, NY USA

来源：

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION | 2024年 / 119卷 / 546期

基金：

欧洲研究理事会;

关键词：

Causal inference; Latent confounding; Model misspecification; Nodewise regression; Structural equation models; INFERENCE; IDENTIFICATION; DISCOVERY; TESTS;

D O I：

10.1080/01621459.2022.2157728

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

We introduce a simple diagnostic test for assessing the overall or partial goodness of fit of a linear causal model with errors being independent of the covariates. In particular, we consider situations where hidden confounding is potentially present. We develop a method and discuss its capability to distinguish between covariates that are confounded with the response by latent variables and those that are not. Thus, we provide a test and methodology for partial goodness of fit. The test is based on comparing a novel higher-order least squares principle with ordinary least squares. In spite of its simplicity, the proposed method is extremely general and is also proven to be valid for high-dimensional settings. for this article are available online.

引用

页码：1019 / 1031

页数：13

共 50 条

[31] Least-squares independence regression for non-linear causal inference under non-Gaussian noise [J].

Makoto Yamada ;

Masashi Sugiyama ;

Jun Sese .

Machine Learning, 2014, 96 :249-267

[32] Least-squares independence regression for non-linear causal inference under non-Gaussian noise [J].

Yamada, Makoto ;

Sugiyama, Masashi ;

Sese, Jun .

MACHINE LEARNING, 2014, 96 (03) :249-267

[33] Higher-order modal transformation for reduced-order modeling of linear systems undergoing global parametric variations [J].

Kim, T. .

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING, 2018, 115 (12) :1477-1498

[34] On the confidence bounds of Gaussian process NARX models and their higher-order frequency response functions [J].

Worden, K. ;

Becker, W. E. ;

Rogers, T. J. ;

Cross, E. J. .

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2018, 104 :188-223

[35] Instrumental Variables Two-Stage Least Squares (2SLS) vs. Maximum Likelihood Structural Equation Modeling of Causal Effects in Linear Regression Models [J].

Maydeu-Olivares, Alberto ;

Shi, Dexin ;

Rosseel, Yves .

STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 2019, 26 (06) :876-892

[36] Estimating hierarchical constructs using consistent partial least squares The case of second-order composites of common factors [J].

van Riel, Allard C. R. ;

Henseler, Jorg ;

Kemeny, Ildiko ;

Sasovova, Zuzana .

INDUSTRIAL MANAGEMENT & DATA SYSTEMS, 2017, 117 (03) :459-477

[37] Asymptotic Behavior of Conditional Least Squares Estimators for Unstable Integer-valued Autoregressive Models of Order 2 [J].

Barczy, Matyas ;

Ispany, Marton ;

Pap, Gyula .

SCANDINAVIAN JOURNAL OF STATISTICS, 2014, 41 (04) :866-892

[38] Extending Regularized Least Squares Support Vector Machines for Order Selection of Dynamical Takagi-Sugeno Models [J].

Kahl, Matthias ;

Kroll, Andreas .

IFAC PAPERSONLINE, 2020, 53 (02) :1182-1187

[39] Logistic regression with peer-group effects via inference in higher-order Ising models [J].

Daskalakis, Constantinos ;

Dikkala, Nishanth ;

Panageas, Ioannis .

INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 108, 2020, 108 :3653-3662

[40] What if we intervene?: Higher-order cross-lagged causal model with interventional approach under observational design [J].

Castro C. ;

Michell K. ;

Kristjanpoller W. ;

Minutolo M.C. .

Neural Computing and Applications, 2024, 36 (24) :15075-15090

← 1 2 3 4 5 →