Pattern Mixture Model with Application in Clinical Trial

来源 :上海交通大学 | 被引量 : 0次 | 上传用户：boboluping

【摘要】

：

Longitudinal data is one of the important primary endpoint types in clinical trial.Usually it is analyzed using mixed-effects model repeated measures(MMRM).

【作者】

：

Shuguang Sun

【机构】

：

Roche Product Development in Asia Pacific,201203,China

【出处】

：

上海交通大学

【发表日期】

：

2016年4期

【关键词】

：

Pattern Mixture Model Missing data

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　Longitudinal data is one of the important primary endpoint types in clinical trial.Usually it is analyzed using mixed-effects model repeated measures(MMRM).

其他文献

Statistical inference for the single index hazards model

In this talk,we develop a class of consistent estimators of the parameter vector for the single-index hazards model,where the model allows nonparametric modeling of covariate effects in a parsimonious

会议

Doubly robustEstimation equationsLocal linear smoothingSemiparametric efficie

TESTING DIMENSIONALITY OF MULTIVARIATE VARIANCE COMPONENTS

Let A be a standard Gaussian random matrix in the space Sym(n)of n n real symmetric(or Hermitian)matrices.The distribution of A is referred to as the GOE(or GUE,respectively).

会议

Chi-bar-square distributiongrowth curvePfaffianspike modelWishart distributi

Estimate treatment effect among completers in clinical trials

会议

Predicting Characteristics of Financial Time Series with Application to the State of Qatar Stock Exc

Forecasting the future is an important issue in time series data analysis in order to aid in planning and the adoption of necessary policies that depend on variables of interest.

会议

Time SeriesForecastingNeural NetworkStock MarketQatar

Ignorability and Unbalanced Longitudinal Data

Longitudinal studies typically propose a schedule of measurement times of participants in advance.Nevertheless,these timings may be affected by some internal(biological conditions of the participants)

会议

longitudinal dataignorabilityinformative observationscovariance modelling

ORTHOGONALIZING EM:A DESIGN-BASED LEAST SQUARES ALGORITHM

We introduce an efficient iterative algorithm,intended for various least squares problems,based on a design of experiments perspective.

会议

Design of experimentsMissing dataOptimizationOrthogonal designThe Lasso

Change-Point Modeling of Copy Number Aberrations in Cancer Genomics

In normal somatic cells,there are two copies of every chromosome.In cancer cells,copy number aberration(CNA)is an important type of genomic change,where large duplications and deletions lead to copy n

会议

change-pointcopy number aberrationscancer genomicsnormalizationsegmentation

A Bayesian Approach for Determining Dose-Response Relationships Based on Multiple Models

Finding appropriate doses of a drug to carry forward is a central issue at every stage of pharmaceutical drug development.Conventional dose-response analyses seek to answer four key questions: Are the

会议

Likelihood PrincipleMCMCConditioning on Data

Multimode data analytics in lung cancer genomi

Despite the major advances in target therapy,lung cancer remains highly deadly.It still leads the cause of cancer mortality worldwide.

会议

Cancer genomicsPatient heterogeneityEGFR mutation

THE UTILITY OF TRACING STUDIES IN COHORTS WITH LOSS TO FOLLOW-UP

Tertiary care facilities often maintain registries of patients with longitudinal follow-up to create a platform for research on the course of chronic disease and the identification of associated risk

会议

dependent observationstatistical efficiencymissing datatracingauxiliary data

Pattern Mixture Model with Application in Clinical Trial

其他学术论文