On Jun 17, 2009, at 1:03 PM, David Winsemius wrote:
On Jun 17, 2009, at 12:35 PM, amor Gandhi wrote:
Hello,
I am using
mod1 <- lrm(y~x1+x2,na.action=na.pass,method="lrm.fit")
summary(mod1)
and I've got the following error:
Error in summary.Design(mod1) : adjustment values not defined here
or with datadist for x1 x2
Several of the useful features of the Design/Hmisc function set
depend on having the datadist option defined. My encounters with
failing to meet that expectation usually occur when I go to plot an
interaction surface with plot.Design(). In my experience some
functions work better if the data is passed in a dataframe. but the
docs for datadist do not imply this to be a requirement.
What do you get if you try:
df1 <- data.frame(y=y, x1=x1, x2=x2) # might not be necessary
dd=datadist(df) # might work with
datadist(y,x1,x2)
^df1^
options(datadist="dd")
mod1 <- lrm(y~x1+x2,na.action=na.pass, data=df1, method="lrm.fit")
# or mod1 <- lrm(y~x1+x2,na.action=na.pass, method="lrm.fit")
summary(mod1)
--
David Winsemius, MD
Heritage Laboratories
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.
