Estimating Survival Treatment Effects with Covariate Adjustment Using Propensity Score

Propensity score is widely used to estimate treatment effects in observational studies.The covariate adjustment using propensity score is the most straightforward method in the literature of causal inference.In this article,we estimate the survival treatment effect with covariate adjustment using propensity score in the semiparametric accelerated failure time model.We establish the asymptotic properties of the proposed estimator by simultaneous estimating equations.We conduct simulation studies to evaluate the finite sample performance of the proposed method.A real data set from the German Breast Cancer Study Group is analyzed to illustrate the proposed method.

A fusion of least squares and empirical likelihood for regression models with a missing binary covariate

Multiply robust inference has attracted much attention recently in the context of missing response data. An estimation procedure is multiply robust, if it can incorporate information from multiple candidate models, and meanwhile the resulting estimator is consistent as long as one of the candidate models is correctly specified. This property is appealing, since it provides the user a flexible modeling strategy with better protection against model misspecification. We explore this attractive property for the regression models with a binary covariate that is missing at random. We start from a reformulation of the celebrated augmented inverse probability weighted estimating equation, and based on this reformulation, we propose a novel combination of the least squares and empirical likelihood to separately handle each of the two types of multiple candidate models,one for the missing variable regression and the other for the missingness mechanism. Due to the separation, all the working models are fused concisely and effectively. The asymptotic normality of our estimator is established through the theory of estimating function with plugged-in nuisance parameter estimates. The finite-sample performance of our procedure is illustrated both through the simulation studies and the analysis of a dementia data collected by the national Alzheimer's coordinating center.

Robust variance estimation for covariate-adjusted unconditional treatment effect in randomized clinical trials with binary outcomes

To improve the precision of estimation and power of testing hypothesis for an unconditional treatment effect in randomized clinical trials with binary outcomes,researchers and regulatory agencies recommend using g computation as a reliable method of covariate adjustment.How-ever,the practical application of g-computation is hindered by the lack of an explicit robust variance formula that can be used for different unconditional treatment effects of interest.To fill this gap,we provide explicit and robust variance estimators for g-computation estimators and demonstrate through simulations that the variance estimators can be reliably applied in practice.

VARIABLE SELECTION FOR COVARIATE ADJUSTED REGRESSION MODEL

This paper employs the SCAD-penalized least squares method to simultaneously select variables and estimate the coefficients for high-dimensional covariate adjusted linear regression models.The distorted variables are assumed to be contaminated with a multiplicative factor that is determined by the value of an unknown function of an observable covariate.The authors show that under some appropriate conditions,the SCAD-penalized least squares estimator has the so called "oracle property".In addition,the authors also suggest a BIC criterion to select the tuning parameter,and show that BIC criterion is able to identify the true model consistently for the covariate adjusted linear regression models.Simulation studies and a real data are used to illustrate the efficiency of the proposed estimation algorithm.

Preface

Multiply robust inference has attracted much attention recently in the context of missing response data. An estimation procedure is multiply robust, if it can incorporate information from multiple candidate models, and meanwhile the resulting estimator is consistent as long as one of the candidate models is correctly specified. This property is appealing, since it provides the user a flexible modeling strategy with better protection against model misspecification. We explore this attractive property for the regression models with a binary covariate that is missing at random. We start from a reformulation of the celebrated augmented inverse probability weighted estimating equation, and based on this reformulation, we propose a novel combination of the least squares and empirical likelihood to separately handle each of the two types of multiple candidate models,one for the missing variable regression and the other for the missingness mechanism. Due to the separation, all the working models are fused concisely and effectively. The asymptotic normality of our estimator is established through the theory of estimating function with plugged-in nuisance parameter estimates. The finite-sample performance of our procedure is illustrated both through the simulation studies and the analysis of a dementia data collected by the national Alzheimer's coordinating center.

On Two-stage Estimate Based on Independent Estimate of Covariance Matrix

When an independent estimate of covariance matrix is available, we often prefer two-stage estimate （TSE）. Expressions of exact covarianee matrix of the TSE obtained by using all and some covariables in eovariance adjustment approach are given, and a necessary and sufficient condition for the TSE to be superior to the least square estimate and related large sample test is also established. Furthermore the TSE, by using some covariables, is expressed as weighted least square estimate. Basing on this fact, a necessary and sufficient condition for the TSE by using some covariables to be superior to the TSE by using all eovariables is obtained. These results give us some insight into the selection of covariables in the TSE and its application.