We consider the smoothed maximum likelihood estimator and the smoothed Grenander-type estimator for a monotone baseline hazard rate 0 in the Cox model. We analyze their asymptotic behaviour and show that they are asymptotically normal at rate nm=.2mC1/, when 0 is m 2 times continuously differentiable, and that both estimators are asymptotically equivalent. Finally, we present numerical results on pointwise confidence intervals that illustrate the comparable behaviour of the two methods.

Original languageEnglish
Pages (from-to)753-791
Number of pages39
JournalScandinavian Journal of Statistics: theory and applications
Publication statusPublished - 2018

    Research areas

  • asymptotic normality, Cox regression model, hazard rate, isotonic estimation, kernel smoothing, smoothed Grenander estimator, smoothed maximum likelihood estimator

ID: 46605061