Bayesian nonparametric generative models for causal inference with missing at random covariates

Autor: Jordan D. Dworkin, Vincent Lo Re, Kirsten J. Lum, Jason Roy, Bret Zeldow, Michael J. Daniels
Rok vydání: 2018
Předmět:
FOS: Computer and information sciences
Statistics and Probability
Biometry
HIV Infections
Bayesian inference
01 natural sciences
Article
General Biochemistry
Genetics and Molecular Biology

Cohort Studies
Methodology (stat.ME)
010104 statistics & probability
03 medical and health sciences
0302 clinical medicine
Joint probability distribution
Statistics
Econometrics
Humans
Statistics::Methodology
Computer Simulation
030212 general & internal medicine
Imputation (statistics)
0101 mathematics
Statistics - Methodology
Mathematics
Models
Statistical

Statistics::Applications
General Immunology and Microbiology
Coinfection
Applied Mathematics
Bayes Theorem
Confounding Factors
Epidemiologic

General Medicine
Missing data
Hepatitis C
Causality
Dirichlet process
Observational Studies as Topic
Causal inference
Parametric model
General Agricultural and Biological Sciences
Algorithms
Quantile
Zdroj: Biometrics
ISSN: 1541-0420
0006-341X
DOI: 10.1111/biom.12875
Popis: We propose a general Bayesian nonparametric (BNP) approach to causal inference in the point treatment setting. The joint distribution of the observed data (outcome, treatment, and confounders) is modeled using an enriched Dirichlet process. The combination of the observed data model and causal assumptions allows us to identify any type of causal effect - differences, ratios, or quantile effects, either marginally or for subpopulations of interest. The proposed BNP model is well-suited for causal inference problems, as it does not require parametric assumptions about the distribution of confounders and naturally leads to a computationally efficient Gibbs sampling algorithm. By flexibly modeling the joint distribution, we are also able to impute (via data augmentation) values for missing covariates within the algorithm under an assumption of ignorable missingness, obviating the need to create separate imputed data sets. This approach for imputing the missing covariates has the additional advantage of guaranteeing congeniality between the imputation model and the analysis model, and because we use a BNP approach, parametric models are avoided for imputation. The performance of the method is assessed using simulation studies. The method is applied to data from a cohort study of human immunodeficiency virus/hepatitis C virus co-infected patients.
Databáze: OpenAIRE
Nepřihlášeným uživatelům se plný text nezobrazuje