Guosheng Yin (The University of Texas M. D. Anderson Cancer Center, USA)

Vendredi 4 avril 2008 à 11h15

The generalized method of moments (GMM) has an attractive structure and is particularly useful to improve estimation efficiency when the likelihood formulation is difficult and the moment conditions are relatively easy to obtain. To enhance the estimation efficiency, we propose the GMM approach to the accelerated failure time model with correlated survival data. We study the semiparametric rank estimator using the martingale-based moments. We circumvent direct estimation of correlation parameters by concatenating the moments and minimizing a quadratic objective function. We establish the consistency and asymptotic normality properties for the parameter estimates, and derive the limiting distribution for the objective function. We carry out simulation studies to examine the finite sample properties of the GMM estimation and inference procedures, and demonstrate its substantial efficiency gain over the conventional method. Finally, we illustrate the new proposal with a real data example from a diabetic retinopathy study. This is joint work with Hui Li.