Semiparametric estimation with data missing not at random using an instrumental variable
From MaRDI portal
Abstract: Missing data occur frequently in empirical studies in health and social sciences, often compromising our ability to make accurate inferences. An outcome is said to be missing not at random (MNAR) if, conditional on the observed variables, the missing data mechanism still depends on the unobserved outcome. In such settings, identification is generally not possible without imposing additional assumptions. Identification is sometimes possible, however, if an instrumental variable (IV) is observed for all subjects which satisfies the exclusion restriction that the IV affects the missingness process without directly influencing the outcome. In this paper, we provide necessary and sufficient conditions for nonparametric identification of the full data distribution under MNAR with the aid of an IV. In addition, we give sufficient identification conditions that are more straightforward to verify in practice. For inference, we focus on estimation of a population outcome mean, for which we develop a suite of semiparametric estimators that extend methods previously developed for data missing at random. Specifically, we propose inverse probability weighted estimation, outcome regression-based estimation and doubly robust estimation of the mean of an outcome subject to MNAR. For illustration, the methods are used to account for selection bias induced by HIV testing refusal in the evaluation of HIV seroprevalence in Mochudi, Botswana, using interviewer characteristics such as gender, age and years of experience as IVs.
Recommendations
- Semiparametric Inference for Nonmonotone Missing-Not-at-Random Data: The No Self-Censoring Model
- A general instrumental variable framework for regression analysis with outcome missing not at random
- Semiparametric maximum likelihood estimation with data missing not at random
- Semiparametric estimating equations inference with nonignorable missing data
- Semiparametric estimation with missing covariates
Cited in
(32)- Multiply robust estimation of causal effects using linked data
- Fully nonparametric inverse probability weighting estimation with nonignorable missing data and its extension to missing quantile regression
- A unified framework of analyzing missing data and variable selection using regularized likelihood
- A general instrumental variable framework for regression analysis with outcome missing not at random
- A novel semiparametric approach to nonignorable missing data by catching covariate marginal information
- Using missing types to improve partial identification with application to a study of HIV prevalence in Malawi
- Instrumental variables estimation with partially missing instruments
- Semiparametric estimation in generalized additive partial linear models with nonignorable nonresponse data
- On semiparametric instrumental variable estimation of average treatment effects through data fusion
- A self-censoring model for multivariate nonignorable nonmonotone missing data
- Treatment effects estimation with missing not at random data without outcome modeling
- Empirical investigations of boosting with pseudo-outcome imputation for missing responses
- Semiparametric Inference for Nonmonotone Missing-Not-at-Random Data: The No Self-Censoring Model
- Double sampling for informatively missing data in electronic health record-based comparative effectiveness research
- Group Testing Regression Analysis with Missing Data and Imperfect Tests
- Semiparametric instrumental variable estimation of simultaneous equation sample selection models
- Boosting Prediction with Data Missing Not at Random
- Causal and counterfactual views of missing data models
- Discussion on: ``Causal and counterfactual views of missing data models
- Response to discussions of: ``Causal and counterfactual views of missing data models
- scientific article; zbMATH DE number 7219858 (Why is no real title available?)
- Shape-restricted statistical inference for non-ignorable missing data under a general additive model
- Testing the missing at random assumption in generalized linear models in the presence of instrumental variables
- On distance functions in multiply robust estimation of population means
- Regression-based imputation of explanatory discrete missing data
- Efficient estimation in a partially specified nonignorable propensity score model
- Nonparametric estimation of path-specific effects in the presence of nonignorable missing covariates
- Semiparametric estimation of the average causal effect of treatment on an outcome measured after a postrandomization event, with missing outcome data
- Causal inference with outcomes truncated by death and missing not at random
- A stableness of resistance model for nonresponse adjustment with callback data
- A Versatile Estimation Procedure Without Estimating the Nonignorable Missingness Mechanism
- Semiparametric Inference of Causal Effect with Nonignorable Missing Confounders
This page was built for publication: Semiparametric estimation with data missing not at random using an instrumental variable
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4558452)
