RE: st: FW: ML for logit/ologit

["Thomas Murray (Department of Economics)" <t.l.murray@bham.ac.uk>]

This makes sense -  rho needs to be defined in the program.  I need rho to converge to a value that maximises the log likelihood.  What I want to do is set a range of possible values for rho (between zero and three specifically) and find a point estimate using ML.  Is this possible to define?

Thanks,

Tom

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: 16 November 2011 11:36
To: 'statalist@hsphsun2.harvard.edu'
Subject: RE: st: FW: ML for logit/ologit

You need to supply the right number of arguments too. Here it looks as if rho is undefined because you are not feeding the program with enough arguments. 

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


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Thomas Murray (Department of Economics)
Sent: 16 November 2011 11:18
To: statalist@hsphsun2.harvard.edu
Subject: RE: st: FW: ML for logit/ologit

Right yes - apologies for that - I was experimenting with different arguments and emailed a faulty program.

What I type is:

program drop logittest
program define logittest
version 12.0
args lnf b1 x rho
tempvar lng
qui {
gen double `lng' = ln(invlogit(`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==1
replace `lng' =  ln(invlogit(-`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==0
replace `lnf' = `lng'
}
end

ml model lf logittest (happy	 =)
ml max

The trace returns:

- version 12.0
          - args lnf b1 x rho
          - tempvar lng
          - qui {
          - gen double `lng' = ln(invlogit(`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==1
          = gen double __000007 = ln(invlogit(__000006*((^(1-)-1)/(1-)))) if happy==1
unknown function ^()
            replace `lng' = ln(invlogit(-`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==0
            replace `lnf' = `lng'

This is Brendan's comment from a bit earlier: 

You have "((^(1-)" in there, so the ^ doesn't apply to a variable, and therefore isn't recognised as an operator.

I think this is because your `x' expands to "".

Could ^ be failing to pick up `rho' inside the bracket?

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nick Cox
Sent: 16 November 2011 11:03
To: 'statalist@hsphsun2.harvard.edu'
Subject: RE: st: FW: ML for logit/ologit

That is not clear at all. The explanation is different. You are invoking a local rho, which you never define. Your earlier statement defines mu. 

It now also seems that your earlier "along the lines of" means that you were not showing us the code that was buggy, which does help to explain why we couldn't see the bug. 

Please do "Say exactly what you typed and exactly what Stata typed (or did) in response."

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

Thomas Murray (Department of Economics)

I have run the set trace and it is clear there is a problem with the including an argument within the squared term (I've pasted the relevant part below).  I am sure this is the bug but I do not know why Stata doesn't like it.

I am confident all the quotations are correct in the program.

Many Thanks,

Tom

- mata: Mopt_search()
        ------------------------------------------------------------------------------------------------ begin logittest ---
        - version 12.0
        - args lnf b1 b2 x mu
        - tempvar lng
        - qui {
        - gen double `lng' = ln(invlogit(`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==1
        = gen double __000007 = ln(invlogit(__000006*((^(1-)-1)/(1-)))) if educ1==1 unknown function ^()
          replace `lng' = ln(invlogit(-`b1'*((`x'^(1-`rho')-1)/(1-`rho')))) if $ML_y1==0
          replace `lnf' = `lng'


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/