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
Many thank,
Amor
[[alternative HTML version deleted]]
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 > > Many thank, > AmorPlease read the documentation for the Design package. Also, type ?lrm Frank> > > > [[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.-- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University
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) 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