st: RE: RE: RE: par ambiguous abbreviation error in qic

[Nick Cox <n.j.cox@durham.ac.uk>]

Yes; the author of the program Jisheng (James) Cui appears to have moved. 

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

Daegyu Yang

Thank you so much! Renaming variables beginning w/ "par*" was the very solution. I really appreciate your kind explanation! 
As you recommend, I also tried to send a mail to the author of the qic command first, but the mail was immediately returned due to the unknown address by the server. So I had to ask here.  Very happy to get the answer this soon. :D

Nick Cox

-qic- is user-written. Please remember to explain where you got user-written programs you refer to. 

SJ-8-1  st0126_1  . . . . . .  QIC program and model selection in GEE analyses
        (help qic if installed) . . . . . . . . . . . . . . . . . . . . J. Cui
        Q1/08   SJ 8(1):146
        general negative binomial distribution now included with qic

SJ-7-2  st0126  . . . . . . .  QIC program and model selection in GEE analyses
        (help qic if installed) . . . . . . . . . . . . . . . . . . . . J. Cui
        Q2/07   SJ 7(2):209--220
        provides the quasilikelihood under the independence
        model criterion (QIC) method for GEE model selection

-qic- estimates a scalar -par-. If you have a variable -par- as well, or one that abbreviates to -par-, there will be a problem. See 

SJ-6-2  dm0021  . Stata tip 31: Scalar or variable? Problem of ambiguous names
        . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  G. I. Kolev
        Q2/06   SJ 6(2):279--280                                 (no commands)
        tips for avoiding abbreviation conflicts with variables
        when naming scalars

and the references it cites for more on this problem. (This paper is accessible online to you, regardless of whether you subscribe to the Stata Journal.)

Thus look to see if you have variables of that kind. 

. d par* 

will tell you. If you have, -rename- such variables as a quick work-around. If that is not the problem, tell us more. 

Daegyu Yang

I tried to use qic in order to know the best correlation structure in my analysis, but the "par ambiguous abbreviation r(111)" error kept occurring. I don't know why. In the statalist, I found that someone had the same problem in 2007 as I do now, but it seems that that person didn't get the solution. Please let me know how to solve this problem or at least why this kind of error keeps occurring.


*
*   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/