[R] Loop and loop output [Cox model, for, function, loglik]

[David Lloyd]

Dear R users,

Below I have written 4 functions CIT1, CIT2a and CIT2b and CIT3 which
recode a variable CLD_ISCH into 3 new variables(T1 T2 T3), I wish to use
T1, T2 and T3 based on the values of tf1 and tf2.
(NOTE:- T2a is used to create T2 in a long winded manner due to my lack
of programming experience)

I then attach T1 T2 and T3 to a dataset KidneyT that contains other
variables i wish to use in a Cox regression model. I require the 
log likelihood statistic of a Cox model when variables T1 T2 T3 and some
from KidneyT are fitted. ( coxmodel$loglik[2] ) 

########################################################################
##########
I can get this to work fine for specified values of tf1 and tf2, but
wish to get the loglikelihood value for many combinations of tf1 and tf2
which are times from 1 to 50 hours i.e I need loglik statistic when tf1
=1 and tf2 = 1 to 50, tf1=2 and tf2=1 to 50. 
I'd like to do this in increments of 0.5 hours.
Ideally I'd like to end up with a matrix/array like

tf1 tf2 loglik
1 0.5 -1200
1 1 -1250
1 1.5 -1251
2 0.5 -1252
2 1 -1249
2 1.5 -1159
etc.

In the hope of finding a maximum for loglik.
########################################################################
##########

I've looked into loops and apply and many other functions but can't seem
to get what i want. Any suggestions would be fantastic - else I'll be
doing hundreds of calculations 1 at a time - ahhhhh!

Code is below: - sorry for not including an example dataset. Am hoping
the code itself. If not pls say and i'll try and repost with a small
example dataset.

#########################################################
tf1=11 
tf2=18 

CIT1=function(var1,bp1){ 

{ 
ifelse (var1<=bp1,(var1=var1),(bp1)) 
} 
} 
T1=CIT1(CLD_ISCH,tf1) 
#-----------------------------------------------------------------------
-------# 
CIT2a=function(var1,var2,bp1){ 

{ 
ifelse (var1<=bp1,(var1=0),(var1=var2-bp1)) 
} 
} 
T2a=CIT2a(CLD_ISCH,CLD_ISCH,tf1) 
#-----------------------------------------------------------------------
-------# 
CIT2b=function(var1,var2,bp1,bp2){ 

{ 
ifelse (var1>(bp2-bp1),(var1=bp2-bp1),(var1=T2a)) 
} 
} 
T2=CIT2b(T2a,T2a,tf1,tf2) 

#-----------------------------------------------------------------------
-------# 

CIT3=function(var1,var2,bp2){ 

{ 
ifelse (var1<=bp2,(var1=0),(var1=var2-bp2)) 
} 
} 
T3=CIT3(CLD_ISCH,CLD_ISCH,tf2) 
#-----------------------------------------------------------------------
-------#
#-----------------------------------------------------------------------
-------#
newdata=data.frame(T1,T2,T3,kidneyT) 
attach(newdata) 

coxmodel=coxph(Surv(TSURV,STATUS)~RAGE+DAGE+REG_WTIME_M+POLY_VS+T1+T2+T3
, 
method="breslow", data=newdata) 

coxmodel$loglik[2] 

##########################################################

Regards and Thx
DaveL



Buying or selling a home? Click here for free info on real estate
services.
<http://tagline.bidsystem.com/fc/Ioyw36XILaXQaXpqHLyCgxKj2HWmH06KAobcQpI
xUGTio1SLafTDjw/> 



<span id=m2wTl><p><font face="Arial, Helvetica, sans-serif" size="2" 
style="font-size:13.5px">_______________________________________________________________<BR>Get
 the Free email that has everyone talking at <a href=http://www.mail2world.com 
target=new>http://www.mail2world.com</a><br>  <font color=#999999>Unlimited 
Email Storage &#150; POP3 &#150; Calendar &#150; SMS &#150; Translator &#150; 
Much More!</font></font></span>
        [[alternative HTML version deleted]]

______________________________________________
R-help@stat.math.ethz.ch 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.