On Oct 4, 2010, at 7:29 AM, 笑啸 wrote:
dear professor:
If this is directed to my posting, I am not a professor.
thank you for your help,witn your help i develop the nomogram
successfully.
after that i want to do the internal validation to the model.i ues
the bootpred to do it,and then i encounter problem again,just like
that.(´íÎóÓÚerror to :complete.cases(x, y, wt) :
²»ÊÇËùÓеIJÎÊý¶¼Ò»Ñù³¤(the length of the
augment was different))
i hope you tell me where is the mistake,and maybe i have chosen the
wrong function.
I think so, at least to the limited extent that I I am understanding
your efforts. It appears you are trying to use boot to do simulation,
but the goal is rather unclear. I think you would get further along
the way if you imported your data into R your data and constructed a
model within the rms system.
None of your functions in this posting are creating a new instance of
dfr$y. They are all depending on the single constructed version of the
data which may be peculiar or it may be exactly typical, but one
cannot tell. Rather than boot() you might get further with
replicate(), but I am not inclined to offer further code in support of
what is an unclear and possibly futile direction.
thank you
turly yours
......
load package 'rms'
ddist <- datadist(dfr)
options(datadist='ddist')
n<-100
set.seed(10)
T.Grade<-factor(0:3,labels=c("G0", "G1", "G2","G3"))
Sex<-factor(sample(0:1, 100, replace=TRUE),labels=c("F","M"))
Smoking<-factor(sample(0:1, 100, replace=TRUE),labels=c("No","yes"))
dfr$L<-with(dfr,0.559*as.numeric(T.Grade)-0.896*as.numeric(Smoking)
+0.92*as.numeric(Sex)-1.338)
dfr$y <- with(dfr, ifelse(runif(n) < plogis(L), 1, 0) )
dfr <- data.frame(T.Grade,Sex,Smoking, L, y)
ddist <- datadist(dfr)
options(datadist='ddist')
f<-lrm(y~T.Grade +Sex+Smoking, data=dfr)
nom<-nomogram(f,fun=function(x)1/(1+exp(-x)),fun.at=c(.01,.05,seq(.
1,.9,by=.2),.9,1),funlabel="Risk of Death")
plot(nom, xfrac=0.45)
load package bootstrap
.................the problem....................
This is not a problem statement. It is merely evidence that you have
some sort of unspecified problem.
theta.fit <- function(dfr,y){lsfit(dfr,y)}
theta.predict <- function(fit,dfr){cbind(1,dfr)%*%fit$coef}
(There is already a predict() function for regular regression problems
and a Predict() function for use with rms fit objects, which if you
explained clearly what you were attempting might be urseful.)
sq.err <- function(y,yhat) { (y-yhat)^2}
It's probably possible to get the sum-squared error for a linear fit
from function lm() by more direct means. I do not understand what the
next line is supposed to be doing, so am unable to offer further
suggestion. Refitting the same model 50 times to the same data seems
an uninformative exercise.
results <- bootpred(x,y,50,theta.fit,theta.predict,err.meas=sq.err)
´íÎóÓÚerror to :complete.cases(x, y, wt) :
²»ÊÇËùÓеIJÎÊý¶¼Ò»Ñù³¤(the length of the
augment was different)
È«¹ú×îµÍ¼Û£¬ÌìÌìÔÚ¼Ò³åÕÕƬ£¬24Сʱ·
¢»õÉÏÃÅ£¡
[[alternative HTML version deleted]]
--
David Winsemius, MD
West Hartford, CT
______________________________________________
R-help@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.
