[R] cox memory

[David Winsemius]

I did not test to see if the design of the data.frame you offered was  
valid input to your model, since you did not provide a means to test  
it within R. I only meant to say that creating a very long vector with  
identical entries should not be necessary if the newdata argument were  
constructed properly. What _is_  needed is for newdata to have all of  
the variables on the RHS of the survfit formula.

Comparing your model, to the newdata argument :

out1<-coxph( Surv(tstart,tstop, death1) ~ chemo1+chemo1:log(tstop 
+1)+horm1+horm1:log(tstop+1)+age1+grade1+grade1:log(tstop 
+1)+positive1+positive1:log(tstop+1)+size1+size1:log(tstop+1), test1)

... I do not for instance see "tstop" defined in newdata. If the  
various survival prediction functions do not get a properly  
constructed newdata argument, they look to the original data values,  
which I believe is what is generating your error. I would think that  
tstop values would needed. It might be useful if you would, at the  
very least, provide the output from str(test1).

Let me also admit that I am unfamiliar with the strategy and  
interpretative implications of including a function of tstop on the  
RHS of a formula that involves tstop as a time variable on the LHS as  
well.

-- 
David

On Sep 22, 2009, at 12:28 PM, Λεωνίδας Μπαντής wrote:

>
> If I use the following
>
> newdata=data.frame(chemo1=0,
> horm1=0,
> age1=mean(age1),
> grade1=0,
> positive1=1,
> size1=mean(size1)   )
>
> then I get
>
> Error in model.frame.default(Terms, newdata, ...) :
>   variable lengths differ (found for 'log(tstop + 1)')
> In addition: Warning message:
> 'newdata' had 1 rows but variable(s) found have 289205 rows
>
>
> and
> that is why I used "rep" to create the profile of a hypothetical  
> subject I want to see.. I've
> seen the help and the web and the syntax for a hypothetical subject is
> indeed as you mentioned, but it is not working (or am I making a
> mistake?). Once you declare newdata as the above then newdata is
> only a row which will conflict with the dimensions of out1. Am I not
> right?
>
> But when I use the rep function to create 289205 zeros for chemo1  
> horm1 and so on I get an out of memory error.
>
> I use Windows XP professional.
>
> Thanx for your answer David.
>
>
>
>
> ___________________________________________________________
> ×ñçóéìïðïéåßôå Yahoo!;
> ÂáñåèÞêáôå ôá åíï÷ëçôéêÜ ìçíýìáôá  
> (spam); Ôï Yahoo! Mail
> äéáèÝôåé ôçí êáëýôåñç äõíáôÞ  
> ðñïóôáóßá êáôÜ ôùí åíï÷ëçôéêþí
> ìçíõìÜôùí http://login.yahoo.com/config/mail?.intl=gr
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.

David Winsemius, MD
Heritage Laboratories
West Hartford, CT