Methods comparing survival functions introductory

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: nctions Introductory example A hypothesis test Special cases 29/42 Actuarial Statistics – Module 2: Non-parametric methods Comparing survival functions Introductory example 1 Introduction 2 Censoring and truncation Censoring Truncation Likelihood for censored and truncated data 3 Estimating the lifetime distribution: non-parametric approach Preliminaries Maximum Likelihood with censoring The likelihood as a function of the discrete hazard function Kaplan-Meier (product limit) estimator Nelson-Aalen estimator 4 Comparing survival functions Introductory example A hypothesis test Special cases 29/42 Actuarial Statistics – Module 2: Non-parametric methods Comparing survival functions Introductory example Based on data representing weeks to death (or censoring) in 51 adults with recurrent gliomas: A=astrocytoma and G=glioblastoma, the following survival functions have been constructed (Example 2.2) Are these two survival functions different? 29/42 Actuarial Statistics – Module 2: Non-parametric methods Comparing survival functions A hypothesis test 1 Introduction 2 Censoring and truncation Censoring Truncation Likelihood for censored and truncated data 3 Estimating the lifetime distribution: non-parametric approach Preliminaries Maximum Likelihood with censoring The likelihood as a function of the discrete hazard function Kaplan-Meier (product limit) estimator Nelson-Aalen estimator 4 Comparing survival functions Introductory example A hypothesis test Special cases 30/42 Actuarial Statistics – Module 2: Non-parametric methods Comparing survival functions A hypothesis test Comparing survival functions In many applications, one wants to compare two populations. For example: smokers versus non-smokers effect of different treatments for a disease As there is a 1-1 relationship between survival function (S ) and hazard rates (µ’s, here denoted h’s), we can test for differences between hazard rates We will test the hypothesis H0 : h1 (t ) = h2 (t ) ; ∀t ≤ τ vs H1 : At least one of the h1 (t ) differ from h2 (t ) for some t ≤ τ . 30/42 Actuarial Statistics – Module 2: Non-parametric methods Comparing survival functions A hypothesis test To test for difference in hazard rates h1 (t ) and h2 (t ) of two different populations for all time t ≤ τ , the general form of the statistic is k dj d1j − , Z1 = w (tj ) n1j nj j =1 where w (tj ) represent a positive weight function t1 < t2 < · · · < tk are the distinct death times in...
View Full Document

This document was uploaded on 04/03/2014.

Ask a homework question - tutors are online