annuncio seminario

[Elena Stanghellini <stanghel>]

Seminari del Dipartimento di Scienze Statistiche, Universita' di Perugia.

Prof. Mounir Mesbah, Direttore del centro SABRES, Università di Vannes
Bretagne 

Venerdi' 31 maggio 2002 alle ore 15:00, Aula Interna del Dipartimento 

Titolo: Joint modeling of a longitudinal latent Quality of Life Variable and 
Event Time  Mounir Mesbah


Abstract:
A major problem that frequently arises in longitudinal clinical trials where 
Quality of Life (QoL) questionnaires are repeatedly assessed by each individual 
under study, is that of dropouts. In some settings, dropouts may depend on 
missing components of the longitudinal process. Dropouts is then termed 
nonignorable. Recently, several approaches have been proposed that accommodate 
nonignorable dropouts in the modeling of a longitudinal process. However, 
these approaches rest upon the assumption that dropouts can only occur at one 
of the pre-specified measurement times of quality of life. In this paper, 
we propose a joint model for the longitudinal latent QoL variable with 
nonignorable dropouts and time to dropout. We obtain parameter estimates in 
this model using nonparametric maximum likelihood estimation. We prove that our 
model is identifiable and that the estimations  are consistent and normally 
distributed.  Finally a Goodness of Fit test for this model is proposed. 

This method is valid when Quality of Life data is just an observed 
continuous univariate score, but in practice Quality of Life data generally 
consist in a multivariate dichotomous or polychotomous observed variable used 
to measure a latent unobserved variable (the Quality of Life).  We propose a 
general way to deal with this kind of data, using a joint modeling of 
measurement of QoL by Rasch Model, Time to dropout by Cox Model and 
Longitudinal process subject to dropout by an AR(1).
 
References:
Dupuy J.-F. et Mesbah M. (2002), Joint Modelling of Event Time and Nonignorable 
    Missing Longitudinal Data, Lifetime Data Analysis. Juin 2002
Jean-François Dupuy, Ion Grama et Mounir Mesbah (2002) Asymptotic theory 
    for the Cox model with missing time-dependent covariate Technical Report 
    UBS-SABRES/RR--2002-02--FR  Mars 2002 
Mohand-Larbi Feddag, Ion Grama et Mounir Mesbah  (2002) Generalized Estimating 
    Equations (GEE) for mixed logistic models. Technical Report 
    UBS-SABRES/RR--2002-03--FR Mars 2002 Submitted
Sneh Gulati, Jean-François Dupuy et Mounir Mesbah (2002) Goodness of fit 
    of a joint model for event time and nonignorable missing longitudinal 
    quality of life data. Technical Report.
    UBS-SABRES/RR--2002-04--FR Avril 2002

                
Per informazioni:

**********************************************
Elena Stanghellini 
Dipartimento di Scienze Statistiche
Via A. Pascoli - C.P. 1315 Succ. 1
06100 Perugia (Italy)

Tel +39 075 5855229 or 5855242
Fax +39 075 43242

email: stanghel@stat.unipg.it
home page: http://www.stat.unipg.it/DSS/elena.html
**********************************************