[R] How to generate a prediction equation for a stratified survival model that was fitted by cph() in Design package

[Yu, Changhong]

Dear Dr. Harrell,

Thanks for response so quickly. Here is example code to illustrate the
problem.

> n <- 1000
> set.seed(731)
> age <- 50 + 12*rnorm(n)
> label(age) <- "Age"
> sex <- factor(sample(c('Male','Female'), n,
+               rep=TRUE, prob=c(.6, .4)))
> cens <- 15*runif(n)
> h <- .02*exp(.04*(age-50)+.8*(sex=='Female'))
> dt <- -log(runif(n))/h
> label(dt) <- 'Follow-up Time'
> e <- ifelse(dt <= cens,1,0)
> dt <- pmin(dt, cens)
> units(dt) <- "Year"
> dat <- data.frame(dt,e,age,sex)
> dd <- datadist(dat)
> options(datadist='dd')
> f1 <- cph( Surv(dt,e) ~ rcs(age,4) + sex, data=dat,x=TRUE, y=TRUE,
surv=TRUE)
> f2 <- cph( Surv(dt,e)~ age + strat(sex), data=dat, x=TRUE,
y=TRUE,surv=TRUE)
> Function(f1) # works well
function(age = 48.800654,sex = "Male") {0.28612987-0.017321951*
age+0.00012586581*pmax(age-29.600456,0)^3-0.00046285671*pmax(age-44.2731
98,0)^3+0.00042111763*pmax(age-53.769262,0)^3-8.4126731e-05*pmax(age-69.
855597,0)^3-0.64448916*(sex=="Male") }
<environment: 0x02c2aef4>
> Function(f2)
Error in Nam[[i]] : subscript out of bounds

I am using Windows R ( version 2.6.0 2007-10-03).

Best regards,

Changhong Yu

-----Original Message-----
From: Frank E Harrell Jr [mailto:f.harrell at vanderbilt.edu] 
Sent: Thursday, December 11, 2008 6:58 PM
To: Yu, Changhong
Cc: r-help at r-project.org
Subject: Re: [R] How to generate a prediction equation for a stratified
survival model that was fitted by cph() in Design package

Yu, Changhong wrote:
> Dear all,
> 
> I used cph() function from Frank harrell's Design package to create a
> survival model, then used functions 'Function' and 'sascode' to
generate
> prediction equation based on the saved survival model. But it failed.
I
> included a stratified variable in the model. If I removed the
> stratification, they were working well. Does that mean that function
> 'Function' doesn't accept a stratified model?

It is supposed to work for that case.  Please send trivial example code 
with data defined in the code that replicates the problem you are 
seeing, and I'll debug.

Frank

> 
> Any thoughts on this will be appreciated. Thanks.
> 
> 
> Changhong
> 
> ===================================
> 
> P Please consider the environment before printing this e-mail
> 
> Cleveland Clinic is ranked one of the top hospitals
> in America by U.S. News & World Report (2008).  
> Visit us online at http://www.clevelandclinic.org for
> a complete listing of our services, staff and
> locations.
> 
> 
> Confidentiality Note:  This message is intended for
use\...{{dropped:13}}
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 


-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt
University



P Please consider the environment before printing this e-mail

Cleveland Clinic is ranked one of the top hospitals
in America by U.S. News & World Report (2008).  
Visit us online at http://www.clevelandclinic.org for
a complete listing of our services, staff and
locations.


Confidentiality Note:  This message is intended for use\...{{dropped:13}}