Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: Weibull MLE, part II


From   <cthompson@dfpm.utah.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Weibull MLE, part II
Date   Thu, 15 Apr 2004 11:13:04 -0600

After I load the program below, I type 
ml model lf weib1 ()
ml check
upon which the following trace & error are returned:
Test 1:  Calling weib1 to check if it computes log likelihood 
and
         does not alter coefficient vector...
         FAILED; weib1 returned error 198.

Here is a trace of its execution:
---------------------------------------------------------------
-> weib1 __000003 __000002
            - `begin'
            = capture noisily version 8.2: weib1 __000003 
__000002
--------------------------------------------------------------- 
begin weib1 
              - version 8.2
              - args lnf leta lgam
              - tempvar p M R
              - quietly {
              - gen double `p' = exp(`lgam')
              = gen double __000004 = exp()
invalid syntax
                gen double `M' = ($ML_y1*exp(-`leta'))^`p'
                gen double `R' = ln($ML_y1)-`leta'
                replace `lnf' = -`M'+$ML_y2*(`lgam'-
`leta'+(`p'-1)*`R')
                }
--------------------------------------------------------------- 
end weib1 ---
            - `end'
            = set trace off
---------------------------------------------------------------
Fix weib1.

If I use the 'cancer' dataset and use the same -ml model- 
command, the same trace & error are returned.  Perhaps I am 
missing something obvious in the argument specification?  
Again, I suspect that my single vector of datapoints may be the 
problem or, perhaps, my confusion regarding proper 
variable/argument specification in the -ml model- command.  
Just to reiterate, I am interested in estimating the theta & 
beta from the Weibull pdf using my small dataset.  As always, 
any input is appreciated!
cheers, clint      

On 15 Apr 2004 at 11:16, Nick Cox wrote:

> As you say, this program looks essentially that in Gould, 
> Pitblado and Sribney (p. 57). I can't see any 
> typos. I wondered if there would be difficulties 
> if `leta' were ever negative in your data, but 
> Stata does the right thing with e.g. exp(--2). 
> So I am reduced to the old advice 
> 
> Say exactly what you typed and exactly what Stata typed 
> (or did) in response. N.B. exactly! If you can, reproduce 
> the error with one of Stata's provided datasets or a simple 
> concocted dataset that you include in your posting. 
> 
> Nick 
> n.j.cox@durham.ac.uk 
> 
> cthompson@dfpm.utah.edu
> 
> > now I am at a loss as to the proper 
> > implementation of the Weibull MLE program supplied by Gould, 
> > Pitblado, & Scribney when I have a single vector of data 
> > (survival time).  The program I'm using is pasted below:
> > program weib1
> >  	version 8.2
> >  	args lnf leta lgam
> >  	tempvar p M R
> >  	quietly {
> >  		gen double `p' = exp(`lgam')
> >  		gen double `M' = ($ML_y1*exp(-`leta'))^`p'
> >  		gen double `R' = ln($ML_y1)-`leta'
> >  		replace `lnf' = -`M'+$ML_y2*(`lgam'-`leta'+(`p'-1)*`R')
> > 	}
> > end
> > 
> > This program executes without problem when used on the 'cancer' 
> > dataset (with some modification to the variable 'drug' of 
> > course) that is accessible from within Stata but terminates in 
> > error on my dataset with 'invalid syntax'.  My question is 
> > this:  what am I overlooking from a programming perspective?  
> > My objective is estimation of theta & beta and -nlcom- will 
> > accomplish this, however, I wish to take the scenic route on 
> > this one thus my desire to do this with Stata's -ml- commands.  
> 
> *
> *   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/


*
*   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