Shape constrained nonparametric estimators of the baseline distribution in Cox proportional hazards model

TL;DR

This paper develops and analyzes shape-constrained nonparametric estimators for the baseline hazard and density in Cox models, proving their consistency and asymptotic properties.

Contribution

It introduces new Grenander-type estimators for monotone baseline hazard and density functions in Cox models, with proven consistency and asymptotic distributions.

Findings

Estimators are strongly consistent.

Estimators are asymptotically equivalent.

Derived their limit distributions at fixed points.

Abstract

We investigate nonparametric estimation of a monotone baseline hazard and a decreasing baseline density within the Cox model. Two estimators of a nondecreasing baseline hazard function are proposed. We derive the nonparametric maximum likelihood estimator and consider a Grenander type estimator, defined as the left-hand slope of the greatest convex minorant of the Breslow estimator. We demonstrate that the two estimators are strong consistent and asymptotically equivalent and derive their common limit distribution at a fixed point. Both estimators of a nonincreasing baseline hazard and their asymptotic properties are acquired in a similar manner. Furthermore, we introduce a Grenander type estimator for a nonincreasing baseline density, defined as the left-hand slope of the least concave majorant of an estimator of the baseline cumulative distribution function, derived from the Breslow estimator. We show that this estimator is strong consistent and derive its asymptotic distribution at a fixed point.