A General Class of Parametric Models for Recurrent Event Data

Autor:	Russell S. Stocker, Edsel A. Peña
Rok vydání:	2007
Předmět:	Statistics and Probability Weak convergence Counting process Applied Mathematics Gaussian Estimator Covariance symbols.namesake Modeling and Simulation Statistics Parametric model symbols Applied mathematics Martingale (probability theory) Parametric statistics Mathematics
Zdroj:	Technometrics. 49:210-221
ISSN:	1537-2723 0040-1706
Popis:	The general class of models proposed by Pena and Hollander for recurrent event data is considered under a fully parametric specification of the baseline hazard rate function and under the two cases where the model does and does not incorporate frailty components. Estimators of model parameters are presented, and their finite and asymptotic properties are ascertained. For the asymptotic properties, the results of Borgan concerning maximum likelihood estimators in counting process models are used to obtain weak convergence to Gaussian distributions of estimators. However, the required regularity conditions are reformulated into conditions involving gap times, which make it more feasible to obtain explicit theoretical expressions of asymptotic covariances. The procedures are applied to fit the general class of models with a parametric baseline hazard rate function to a dataset on hydraulic subsystems of “load-haul-dump” machines in mining.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::ad2ef33c54c391e2d05165f2e54427bf https://doi.org/10.1198/004017007000000056 Zobrazit plný text záznamu