Estimation and Efficiency with Recurrent Event Data under Informative Monitoring


This article deals with studies that monitor occurrences of a recurrent event for n subjects or experimental units. It is assumed that the ith unit is monitored over a random period [0, τi]. The successive inter-event times Ti 1, Ti 2, ..., are assumed independent of τi. The random number of event occurrences over the monitoring period is Ki = max { k ∈ { 0, 1, 2, ... } : Ti 1 + Ti 2 + ⋯ + Tik ≤ τi }. The Tij's are assumed to be i.i.d. from an unknown distribution function F which belongs to a parametric family of distributions C = { F (· ; θ) : θ ∈ Θ ⊂ Rp }. The τi's are assumed to be i.i.d. from an unknown distribution function G. The problem of estimating θ, and consequently the distribution F, is considered under the assumption that the τi's are informative about the inter-event distribution. Specifically, 1 - G = (1 - F)β for some unknown β > 0, a generalized Koziol-Green [cf., Koziol, J., Green, S., 1976. A Cramer-von Mises statistic for randomly censored data. Biometrika 63, 139-156; Chen, Y., Hollander, M., Langberg, N., 1982. Small-sample results for the Kaplan-Meier estimator. J. Amer. Statist. Assoc. 77, 141-144] model. Asymptotic properties of estimators of θ, β, and F are presented. Efficiencies of estimators of θ and F are ascertained relative to estimators which ignore the informative monitoring aspect. These comparisons reveal the gain in efficiency when the informative structure of the model is exploited. Concrete demonstrations were performed for F exponential and a two-parameter Weibull.


Mathematics and Statistics


National Institute of Health (U.S.)
National Science Foundation (U.S.)

Keywords and Phrases

Weibull inter-event times; counting processes; efficiency comparisons; exponential inter-event times; generalized Koziol-Green model; martingales

Document Type

Article - Journal

Document Version


File Type





© 2010 Elsevier, All rights reserved.