Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Weighted least-squares estimation for Aalen's additive risk model. Technical report

Technical Report ·
OSTI ID:5318889

Cox's (1972) proportional hazards model has so far been the most popular model for the regression of analysis censored survival data. However, the additive risk model of Aalen (1980) 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},...,Y{sub p})', has the form h(t/Y) = Y' {alpha} (t), where {alpha} = ({alpha}{sub 1},...,{alpha}{sub p})' is an unknown vector of hazard functions. This paper discusses inference for alpha 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.

Research Organization:
Stanford Univ., CA (United States). Dept. of Statistics
OSTI ID:
5318889
Report Number(s):
AD-A-235250/8/XAB; TR--441; CNN: N00014-89-J-1627
Country of Publication:
United States
Language:
English