Web-based Supplementary Materials for: “Semiparametric approach for non-monotone missing covariates in a parametric regression model”
نویسندگان
چکیده
Here the simulation design is the same as that in scenario 1 described in the main text with two partially missing variables X1 and X2. Missing data were created by following the two non-ignorable mechanisms, 1) logit{pr(Rk = 1|X1, X2, Y, Z)} = 0.25Y + 0.25Z + X2 + X1 and 2) logit{pr(Rk = 1|X1, X2, Y, Z)} = 0.75 + Y + 0.25Z −X1 + X2, for k = 1, 2. Although in both mechanisms Rk strongly depends on both X1 and X2, dependence on Y is weak and strong for mechanisms 1 and 2, respectively. Also, both mechanisms resulted in approximately 25% missing data for X1 and for X2. The results in Table W-1 show that the complete case method has significant bias in the parameter estimates. As expected, compared to the meanscore approach, the SP method shows much less bias in the estimates. The reason is that the SP method assumes NImechanism which allows R1 to depend on X2 along with Y and Z, and R2 to depend on X1 along with Y and Z – a relatively close model to the true missing mechanism than the MAR mechanism where Rk is assume not to depend on X1 or X2, for k = 1, 2. Note that for all methods, the bias also depends on how strongly the missingness mechanism depends on the response Y .
منابع مشابه
Semiparametric approach for non-monotone missing covariates in a parametric regression model.
Missing covariate data often arise in biomedical studies, and analysis of such data that ignores subjects with incomplete information may lead to inefficient and possibly biased estimates. A great deal of attention has been paid to handling a single missing covariate or a monotone pattern of missing data when the missingness mechanism is missing at random. In this article, we propose a semipara...
متن کاملA Bayesian Semiparametric Model for Case - ControlStudies with Errors
We develop a model and a numerical estimation scheme for a Bayesian approach to inference in case-control studies with errors in covariables. The model proposed in this paper is based on a nonparametric model for the unknown joint distribution for the missing data, the observed covariates and the proxy. This nonparametric distribution deenes the measurement error component of the model which re...
متن کاملWeb - based Supplementary Materials for ( Adjustment for Missingness Using Auxiliary Information in Semiparametric Regression ) by ( Donglin Zeng and Qingxia Chen ) 1 Web
متن کامل
Non-parametric and semiparametric models for missing covariates in parametric regression Abstracts
s Robustness of covariate modeling for the missing covariate problem in parametric regression is studied under the MAR assumption. For a simple missing covariate pattern, non-parametric likelihood is proposed and is shown to yield a consistent and semiparametrically efficient estimator for the regression parameter. Total robustness is achieved in this situation. For more general missing covaria...
متن کاملNon-parametric and semiparametric models for missing covariates in parametric regression Abstracts
s Robustness of covariate modeling for the missing covariate problem in parametric regression is studied under the MAR assumption. For a simple missing covariate pattern, non-parametric likelihood is proposed and is shown to yield a consistent and semiparametrically efficient estimator for the regression parameter. Total robustness is achieved in this situation. For more general missing covaria...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2013