R help - Jul 2008 - package segmented problem

Hi,  while using package "segmented" (version 0.2-4) by Vita Muggeo to
investigate a possible change point (around time = 222) in admissions  
for a specific medical condition I had the following error message:

fit2.seg<-segmented(fit2, seg.Z=~time,psi=222)
Error in segmented.lm(fit2, seg.Z = ~time, psi = 222) :
   (Some) estimated psi out of its range

"fit2" is a simple lm fit to time.  I have used this package  
successfully for related analyses without this error.  Having tried  
many alternatives to "psi=222" but with the same error, I wondered if
anyone can assist?
Many thanks,
Alan Kelly

Hi Alan

How your lm model looks like?
Are your data stored on a data.frame ? Case yes, send us a str(df).
Send us a short sample of the code and a short subset of the data.

Cheers,

miltinho astronauta
brazil

On Tue, Jul 8, 2008 at 5:36 AM, Alan Kelly <akelly@tcd.ie> wrote:
> Hi,  while using package "segmented" (version 0.2-4) by Vita
Muggeo to
> investigate a possible change point (around time = 222) in admissions for a
> specific medical condition I had the following error message:
>
> fit2.seg<-segmented(fit2, seg.Z=~time,psi=222)
> Error in segmented.lm(fit2, seg.Z = ~time, psi = 222) :
>  (Some) estimated psi out of its range
>
> "fit2" is a simple lm fit to time.  I have used this package
successfully
> for related analyses without this error.  Having tried many alternatives to
> "psi=222" but with the same error, I wondered if anyone can
assist?
> Many thanks,
> Alan Kelly
>
> ______________________________________________
> 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<http://www.r-project.org/posting-guide.html>
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]