In many cases, the treatment of missing data in an analysis is carried out in a casual. Author links open overlay panel shigeyuki hamori a. The pretestposttest study is commonplace in numerous appli. Missing data occur frequently in empirical studies in health and social sciences, often compromising our ability to make accurate inferences. This paper considers the problem of parameter estimation in a general class of semiparametric models when observations are subject to missingness at random. Thus, our method, albeit semiparametric in spirit, is totally different from the existing semiparametric imputation methods of, for example, lipsitz et al. Missing data models a search on mathscinet in early may 2005 for semiparametric and missing data gave 15 hits. International conference on robust statistics 2016 1. A semiparametric mixture regression model for longitudinal data authors. In statistics, a semiparametric model is a statistical model that has parametric and nonparametric components.
We consider the efficiency bound for the estimation of the parameters of semiparametric models defined solely by restrictions on the means of a vector of correlated outcomes, y, when the data on y are missing at random. Pdf asymptotic theory for the semiparametric accelerated. Munich personal repec archive semiparametric spatial regression. A major development in this area was the systematic development of information bounds for semiparametric regression models with covariates missing at random by robins, rotnitzky, and. Semiparametric models allow at least part of the data generating process to be. Missing data is a pervasive problem in data analyses, resulting in datasets that contain censored realizations of a target distribution. Semiparametric bayesian analysis of matched casecontrol. Semiparametric theory and missing data by tsiatis, a. A semiparametric mixture regression model for longitudinal.
Pdf analysis of semiparametric regression models for. Asymptotic theory for the semiparametric accelerated failure time model with missing data by bin nan,1 johnd. Kalbfleisch and menggang yu university of michigan, university of michigan and indiana university we consider a class of doubly weighted rankbased estimating methods for the transformation or accelerated failure time model with. Semiparametric theory and missing data researchgate.
Abstract we develop inference tools in a semiparametric partially linear regression model with missing response data. A semiparametric inference to regression analysis with. They can be viewed as an extension of generalized estimating equations estimators that allow for the data to be missing at random but not missing completely at random. Semiparametric bayesian analysis of matched casecontrol studies with missing exposure samiran s inha, bhramar m ukherjee,malayghosh, bani k.
Pdf we consider the efficiency bound for the estimation of the parameters of. The supplementary material gives the semiparametric efficiency theory for estimation of natural direct effects with a known model for the mediator density. We consider a class of doubly weighted rankbased estimating methods for the transformation or accelerated failure time model with missing data as arise, for. Asymptotic theory for the semiparametric accelerated failure time model with missing data. Semiparametric theory and missing data springer series in statistics 9780387324487. The goal is to combine the simplicity of imputation. A semiparametric estimation of mean functionals with. In this book, tsiatis very carefully and didactically explains this theory. Important examples include weighted estimating equations for missing data robins et al. Pdf missing data imputation is an important issue in machine learning and data mining. Estimation in semiparametric models with missing data 789 from the imputed estimating function gn. While our largesample theory allows for a wide range of.
Calibration estimation of semiparametric copula models with data missing at random. A semiparametric inference to regression analysis with missing coariatesv in survey data shu angy and jae kwang kim department of statistics, iowa state university abstract. Calibration estimation of semiparametric copula models. Semiparametric theory and missing data springerlink. Multiple imputation for missing values through conditional. Kalbfleischand menggangyu university of michigan, university of michigan and indiana university we consider a class of doubly weighted rankbased estimating methods for the transformation or accelerated failure time model. Conditional moment models with data missing at random. C arroll this article considers bayesian analysis of matched casecontrol problems when one of the covariates is partially missing. These estimators can be used to correct for dependent. Missing data often appear as a practical problem while applying classical models in the statistical analysis. But for monotone missing data, we propose an adaptive estimator whose. Estimation in semiparametric models with missing data.
Pdf semiparametric efficiency in multivariate regression models. In this paper, we investigate general moment or conditional moment restriction models with missing data and establish an equivalence result whose main message is. Identification, doubly robust estimation, and semiparametric. The description of the theory of estimation for semiparametric models is both rigorous and intuitive, relying on geometric ideas to reinforce the intuition and understanding of the theory. Semiparametric theory and missing data springer series in statistics anastasios tsiatis. The theory of missing data applied to semiparametric models is scattered. Semiparametric regression analysis with missing response. In this paper, we consider a semiparametric regression model in the presence of missing covariates for nonparametric components under a bayesian framework. This paper investigates a class of estimation problems of the semiparametric model with missing data. Zhao 1994 and robins and rotnitzky 1992 are revisited for semiparametric regression models with missing data using the theory outlined in the monograph by bickel, klaassen, ritov, and wellner 1993.
Statistics in the pharmaceutical industry, 3rd edition. In the 90s, jamie robins and colleagues in harvard applied recently developed theory for semiparametric models to the problem of handling missing data. Pdf semiparametric estimation with data missing not at. Influence functions ifs are a core component of classic statistical theory. A statistical model is a parameterized family of distributions. Joint modeling of missing data due to nonparticipation. Further methodology and theory was developed by, e. Analysis of semiparametric regression models for repeated outcomes in the presence of missing data james m.
Strategies for bayesian modeling and sensitivity analysis m. Semiparametric methods for missing data and causal. By adopting nonparametric components for the model, the estimation method can be made robust. Semiparametric theory and missing data anastasios tsiatis. Semiparametric theory and missing data pdf free download. The main results are given in a more relevant format for. This sensitivity is exacerbated when inverse probability weighting methods are used, which may overweight contaminated observations. An application to the 2014 cps asec jonathan rothbaum september 2, 2015 sehsd working paper 201515 abstract the current population survey annual social and economic supplement cps asec serves as the data source for official income, poverty, and inequality statistics in the united states. Missing data in populationbased studies can occur for several reasons, but the most common reasons in studies of older adults are nonparticipation and death. Analysis of semiparametric regression models for repeated. We show that the semiparametric variance bound is the asymptotic variance of the optimal estimator in a class of inverse probability of censoring weighted estimators and that this bound is unchanged if the data are missing. For many semiparametric problems, the estimation of regression coe cients without the task of variable selection does not pertain to the minimization of any objective function. Analysis of semiparametric regression models for repeated outcomes in the presence of missing data march 1995 journal of the american statistical association 90429. Missing data are a common occurrence and can have a significant effect on the conclusions that can be drawn from the data.
This book combines much of what is known in regard to the theory of estimation for semiparametric models with missing data in an organized and comprehensive manner. In statistics, missing data, or missing values, occur when no data value is stored for the variable in an observation. The results of this paper can be viewed as a step in lling the large gap be. Pdf semiparametric optimization for missing data imputation. Peisong han, lu wang, estimation with missing data. Semiparametric theory and missing data, by tsiatis, 2006, 404 pages. An outcome is said to be missing not at random mnar if, conditional on the observed variables, the missing data mechanism still depends on. In any longitudinal study, estimation of the mean of an outcome variable as a function of time is compromised due to missing data 1, 2, 3, 4. Semiparametric statistical theory, which is broadly speaking about the estimation of. The theory of missing data applied to semiparametric models is scattered throughout the literature with no thorough comprehensive treatment of the subject. We may combine both ra and ipw estimators to form a doublyrobust.
A parametric model is a model in which the indexing parameter. Semiparametric efficiency in multivariate regression. The estimation procedure fully utilizes the entire dataset to achieve increased efficiency, and the resulting coefficient estimators are rootn consistent and asymptotically normal. Missing data occurs frequently in empirical studies in health and social sciences, often compromising our ability to make accurate inferences.
Kriging regression imputation method to semiparametric. A semiparametric logistic regression model is assumed for the response probability and a nonparametric regression approach for missing data discussed in cheng 1994 is used in the estimator. Theory and method analysis of semiparametric regression models for repeated outcomes in the presence of missing data. Supplemental appendix to semiparametric theory for causal mediation analysis. Parameter estimation in parametric regression models with missing coariatesv is considered under a survey sampling setup. Classical semiparametric inference with missing outcome data is not robust to contamination of the observed data and a single observation can have arbitrarily large influence on estimation of a parameter of interest. Semiparametric estimation of treatment effect in a pretestposttest study with missing data marie davidian, anastasios a. L1 theory was established by carbon, hallin and tran 1996. Wellner springer verlag this book is a reprint of the book that appeared with johns hopkins university press in 1993. Efficient and adaptive estimation for semiparametric. Penalized estimating functions and variable selection in. See kenward and carpenter 2007 for a comparison of the relative.
In order to overcome the robust defect of traditional complete data estimation method and regression imputation estimation technique, we propose a modified imputation estimation approach called krigingregression imputation. Bridging a survey redesign using multiple imputation. This paper investigates the estimation of semiparametric copula models with data missing at random. Tsiatis, 2006 and the buckleyjames 1979 estimator for semiparametric 2. Efficient and adaptive estimation for semiparametric models 84 9780387984735. Semiparametric methods for missing data and causal inference abstract in this dissertation, we propose methodology to account for missing data as well as a strategy to account for outcome heterogeneity. Statistical science semiparametric estimation of treatment. Springer verlag does the statistical community a great. Asymptotic theory for the semiparametric accelerated.
122 729 296 45 826 248 1263 924 812 505 1364 1460 215 1205 1488 1034 1476 597 280 1252 27 720 1150 784 40 405 355 46 1018 586 1628 651 187 309 185 277 1247 792 1517 118 1323 1217 1055 1468 1354 660