中文  |  English

University of MichiganProf.Yi Li:A Pairwise-Likelihood Augmented Estimator for Cox Model Under Left-Truncation

source:Statistics School of SWUFE Release date:2015-07-13 Views:509

ThemeA Pairwise-Likelihood Augmented Estimator for Cox Model Under Left-Truncation

SpeakerProf.Yi Li

Hosted ByProf.Huazhen Lin

Time PM14:00-15:00, July 142015

Place Academic Meeting Room ( B212 inTongbo Building )

OrganizersStatistical Research Center;School of Statistics;Scientific Bureau

 

Resume

Yi Li , A professor at the university of Michiganwho is the director of Kidney Epidemiology and Cost Center and also an evaluation expert of NSF Statistics and Probability Programis even  an associate editor of five periodicals including Journal of the American Statistical AssociationBiometricsScandinavian Journal of Statistics.Professor Li who engages in the study of statistics and biological statistics ,has presided over 10 funds including the United States national cancer institute, totaling more than $24 million and has published more than 100 papers, about 20 of which are published in the international top journals  such as Journal of the American Statistical AssociationBiometricsBiometrikaJournal of the Royal Statistical Society:Series B. He is a member ofInternational society for biological statisticsASA (about three over one thousand of the American statistical association received this honor )international institute of mathematical statistics and International Chinese Statistical Association

Summary

Survival data collected from prevalent cohorts are subject to left-truncation. Conventional conditional approaches can be inefficient by ignoring the information in the marginal likelihood of the truncation time. On the other hand, the stationarity assumption under length-biased sampling methods will lead to biased estimation when it is violated. In this paper, we propose a semiparametric estimation method by augmenting the Cox partial likelihood with a pairwise likelihood. We eliminate the unspecified truncation distribution in the marginal likelihood, yet retain the information about regression coefficients and the baseline hazard. Self-consistency of the estimator guarantees a fast algorithm to solve for the regression coefficients and thebaseline hazard simultaneously. The proposed estimator is shown to be consistent and asymptotically normal with a consistent sandwich-type variance estimator. Simulations show a substantial efficiency gain in both the regression coefficients and the cumulative hazard over Cox estimators, and that the gain is comparable to LBS methods when the uniform truncation assumption holds. A data analysis illustrates the application of the methods.


2
Share to: