eortc: Simulated data set based on an EORTC trial

Description Usage Format Details Source References Examples

Description

This is a simulated surival data set for investigating random center effects. To make it realistic, the number of centers and their sizes is based on an EORTC cancer trial.

Usage

1

Format

A data frame with 2323 observations on the following 4 variables.

y

survival time

uncens

0=alive, 1=dead

center

enrolling center, a number from 1 to 37

trt

treatment arm, 0 or 1

Details

This is used in the test suite for the code.

Source

PhD thesis work of Jose Cortinas Abrahantes

References

Cortinas Abrahantes, Jose; Burzykowski, Tomasz (2002), A version of the EM algorithm for proportional hazards models with random effects , Published in: Lecture Notes of the ICB Seminars. p. 15-20

Examples

1
2
data(eortc)
coxme(Surv(y, uncens) ~ trt + (trt| center) + strata(center), eortc)

Example output

Loading required package: survival
Loading required package: bdsmatrix

Attaching package: 'bdsmatrix'

The following object is masked from 'package:base':

    backsolve

Cox mixed-effects model fit by maximum likelihood
  Data: eortc
  events, n = 1463, 2323
  Iterations= 5 28 
                    NULL Integrated   Fitted
Log-likelihood -5609.797   -5539.43 -5528.94

                   Chisq    df p    AIC    BIC
Integrated loglik 140.74  2.00 0 136.74 126.16
 Penalized loglik 161.72 10.22 0 141.28  87.25

Model:  Surv(y, uncens) ~ trt + (trt | center) + strata(center) 
Fixed coefficients
         coef exp(coef)  se(coef)    z p
trt 0.7431879  2.102628 0.0784898 9.47 0

Random effects
 Group  Variable Std Dev    Variance  
 center trt      0.24220971 0.05866554

coxme documentation built on May 29, 2017, 5:26 p.m.