# phreg: Fast Cox PH regression In mets: Analysis of Multivariate Event Times

## Description

Fast Cox PH regression

## Usage

 1 phreg(formula, data, ...)

## Arguments

 formula formula with 'Surv' outcome (see coxph) data data frame ... Additional arguments to lower level funtions

Klaus K. Holst

## Examples

 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 simcox <- function(n=1000, seed=1, beta=c(1,1), entry=TRUE) { if (!is.null(seed)) set.seed(seed) library(lava) m <- lvm() regression(m,T~X1+X2) <- beta distribution(m,~T+C) <- coxWeibull.lvm(scale=1/100) distribution(m,~entry) <- coxWeibull.lvm(scale=1/10) m <- eventTime(m,time~min(T,C=0),"status") d <- sim(m,n); if (!entry) d\$entry <- 0 else d <- subset(d, time>entry,select=-c(T,C)) return(d) } ## Not run: n <- 10; d <- mets:::simCox(n); d\$id <- seq(nrow(d)); d\$group <- factor(rbinom(nrow(d),1,0.5)) (m1 <- phreg(Surv(entry,time,status)~X1+X2,data=d)) (m2 <- coxph(Surv(entry,time,status)~X1+X2+cluster(id),data=d)) (coef(m3 <-cox.aalen(Surv(entry,time,status)~prop(X1)+prop(X2),data=d))) (m1b <- phreg(Surv(entry,time,status)~X1+X2+strata(group),data=d)) (m2b <- coxph(Surv(entry,time,status)~X1+X2+cluster(id)+strata(group),data=d)) (coef(m3b <-cox.aalen(Surv(entry,time,status)~-1+group+prop(X1)+prop(X2),data=d))) m <- phreg(Surv(entry,time,status)~X1*X2+strata(group)+cluster(id),data=d) m plot(m,ylim=c(0,1)) ## End(Not run)

mets documentation built on May 30, 2017, 7:43 a.m.