Weighted least squares estimation for Aalen’s additive risk model

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

117 Scopus Citations
View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Pages (from-to)114-129
Journal / PublicationJournal of the American Statistical Association
Volume86
Issue number413
Publication statusPublished - Mar 1991
Externally publishedYes

Abstract

Cox’s proportional hazards model has so far been the most popular model for the regression analysis of censored survival data. However, the additive risk model of Aalen can provide a useful and biologically more plausible alternative. Aalen’s model stipulates that the conditional hazard function for a subject, whose covariates are Y = (Y<sub>1</sub>, …, Y<sub>p</sub>)’, has the form h(t ∣ Y) = Y’ α(t), where α = (α<sub>1</sub>, …, α<sub>p</sub>)’ is an unknown vector of hazard functions. This article discusses inference for α based on a weighted least squares (WLS) estimator of the vector of cumulative hazard functions. The asymptotic distribution of the WLS estimator is derived and used to obtain confidence intervals and bands for the cumulative hazard functions. Both a grouped data and a continuous data version of the estimator are examined. An extensive simulation study is carried out. The method is applied to grouped data on the incidence of cancer mortality among Japanese atomic bomb survivors. © 1991 Taylor & Francis Group, LLC.

Research Area(s)

  • Atomic bomb survivors, Grouped survival data, Martingale methods, MonteCarlo, Multivariate counting processes, Nonproportional hazards

Bibliographic Note

Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to lbscholars@cityu.edu.hk.