Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: clad error: sample size smaller than number of degrees of freedom


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: clad error: sample size smaller than number of degrees of freedom
Date   Wed, 29 Aug 2007 17:19:37 +0100

As I understand it, -clad- slashes at your data until
it gets subsamples that suits its purpose. What 
number you start off with is not so crucial as what 
subsample size it is working with. 

As getting a reasonable model fit is more important
than getting the standard errors right, Vivian might 
like to try her data on -qreg- first. 

Nick 
n.j.cox@durham.ac.uk 

Scott Merryman
 
> I believe it indicates that all of the predict y_hats are less than
> the lower limit (assuming left censoring).  You will need to increase
> the quantile.

Vivian Hoffmann

> > I'm trying to estimate a model using the <clad> command, 
> but I get the
> > following error:
> >
> > Program did not converge,
> > trimmed sample size is smaller than number of degrees of freedom.
> >
> > even though I have 188 observations (98 non-censored) and only one
> > independent variable...
> >
> > Any ideas on why this might be happening or how to solve it?

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index