Statistical Methods for the Analysis of Survival Data

Acronym

P_SURVSTAT

Consortium Coordinator

Sal Y Rosas Celi, Victor Giancarlo

Start Date

March 1, 2014

End Date

February 28, 2015

Status

https://purl.org/pe-repo/concytec/estadoProyecto#concluido

Tipo de proyecto

https://purl.org/pe-repo/ocde/tipoProyecto#investigacionBasica

Description

Advancements in the field of medicine and biology have created new opportunities and challenges for statisticians in the form of new data structures. One of those structures is interval censored data type I that arises when a failure time 'T' can not be observed, but can only be determined to lie before or after an examination times 'Y'. A fundamental problem of interest is to estimate the distribution function of 'T' when 'T' and 'Y' are not independent (also called informative censoring) and when the indicator variable that measures whether or not 'T' occurs before 'Y' is misclassified. With this funding we plan to focus on developing copula and non-parametric statistical models to properly estimate the distribution function of 'T' with this type of data structure. The emphasis will be on methods that are computationally feasible in order to make it available to applied scientists. A number of approaches have been proposed to control for informative censoring in the case of interval censored data type I also know as current status data including sensitive analysis, copula and non-parametric models. In the case of outcome misclassification, the applicant recently developed methods to properly adjust for this problem in the case of interval censored data type I. We proposed to develop statistical methods to 1) estimate the cumulative distribution function of 'T' (one sample problem), 2) propose statistics to test the difference in the cumulative distribution functions between two groups, and 3) propose regression models to measure the effect of several factors on 'T'.

Keywords

Análisis de Supervivencia

;

Estadística Biomédica

;

Datos Censurados

;

Modelos Estadísticos

Área de conocimiento

Natural sciences

;

Engineering and technology

Campo OCDE

https://purl.org/pe-repo/ocde/ford#1.01.03

;

https://purl.org/pe-repo/ocde/ford#2.06.01