Analyse multidimentionnelle de durées de vie

分析人生的多维度

• 批准号: 341760-2007
• 负责人: LakhalChaieb, MHamedLajmi
• 金额: $ 0.87万
• 依托单位: Université Laval
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2007
• 资助国家: 加拿大
• 起止时间: 2007-01-01 至 2008-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=363360
• 关键词: Analyse; multidimentionnelle; de; dur; es

My methodological research interests are in multivariate survival analysis. It consists on the analysis of recurrent events and the analysis of correlated time-to-event variables.In recurrent events analysis, my interest focus on dependent censoring. We are developing methodologies for modelling dependent censoring and estimating the mean function under these conditions. Multistates models and IPCW estimators are investigated.In the analysis of correlated time-to-event variables, my research interest focus on deriving goodness-of-fit procedures to test the adequacy of semiparametric models (Copula, Cox model, ...etc) to assess incomplete data sets. Typically, specific procedures are required for each scheme of incomplete data. Examples of such data include bivariate censored data and data subject to dependent truncation.Other points of interest include nonparametric measure of association between two time-to-events variables with incomplete data and the applications of the Graphic-Copula estimators

我的方法学研究兴趣是多变量生存分析。在复发事件分析中，我的兴趣集中在相依删失。我们正在开发的方法，在这些条件下，建模相关删失和估计的平均函数。本文研究了多状态模型和IPCW估计，在相关时间变量的分析中，主要研究了半参数模型（Copula，考克斯模型等）的拟合优度检验方法。等），以评估不完整的数据集。一般来说，每一个不完整数据的方案都需要特定的程序。这类数据的例子包括双变量删失数据和相依截断数据，其他兴趣点包括不完全数据下两个时间事件变量之间关联的非参数度量和Graphic-Copula估计的应用