2008/5/29 Frank E Harrell Jr [EMAIL PROTECTED]:
Patrizio Frederic wrote:
dear Harrell,
thank you for quick reply and suggestions. I still have the problem:
library(Design)
x = rnorm(100)
y = runif(100)(exp(x)/(1+exp(x)))
y = 0*y+1*y
d = datadist(x,y)
options(datadist=d)
fit =
Patrizio Frederic wrote:
2008/5/29 Frank E Harrell Jr [EMAIL PROTECTED]:
Patrizio Frederic wrote:
dear Harrell,
thank you for quick reply and suggestions. I still have the problem:
library(Design)
x = rnorm(100)
y = runif(100)(exp(x)/(1+exp(x)))
y = 0*y+1*y
d = datadist(x,y)
Thanks for the solution. But I wish someone would tell us why the solution
makes sense. If you have changed your R environment in any way from the
defaults please let us know.
Frank
--
Frank E Harrell Jr Professor and Chair School of Medicine
Department
Patrizio Frederic wrote:
Thanks for the solution. But I wish someone would tell us why the solution
makes sense. If you have changed your R environment in any way from the
defaults please let us know.
Frank
--
Frank E Harrell Jr Professor and Chair School of Medicine
On Fri, 30 May 2008, Patrizio Frederic wrote:
Thanks for the solution. But I wish someone would tell us why the solution
makes sense. If you have changed your R environment in any way from the
defaults please let us know.
Frank
--
Frank E Harrell Jr Professor and Chair School of
Prof Brian Ripley wrote:
On Fri, 30 May 2008, Patrizio Frederic wrote:
Thanks for the solution. But I wish someone would tell us why the
solution
makes sense. If you have changed your R environment in any way from the
defaults please let us know.
Frank
--
Frank E Harrell Jr Professor
On Fri, 30 May 2008, Frank E Harrell Jr wrote:
Prof Brian Ripley wrote:
On Fri, 30 May 2008, Patrizio Frederic wrote:
Thanks for the solution. But I wish someone would tell us why the
solution
makes sense. If you have changed your R environment in any way from the
defaults please let us
dear Harrell,
thank you for quick reply and suggestions. I still have the problem:
library(Design)
x = rnorm(100)
y = runif(100)(exp(x)/(1+exp(x)))
y = 0*y+1*y
d = datadist(x,y)
options(datadist=d)
fit = lrm(y~x)
# works fine, but
plot(fit) #produce the error message
Error in
Patrizio Frederic wrote:
dear Harrell,
thank you for quick reply and suggestions. I still have the problem:
library(Design)
x = rnorm(100)
y = runif(100)(exp(x)/(1+exp(x)))
y = 0*y+1*y
d = datadist(x,y)
options(datadist=d)
fit = lrm(y~x)
# works fine, but
plot(fit) #produce the error
9 matches
Mail list logo
