RE: Antwort: AW: Antwort: st: AW: your Stata exectuable is out of date

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

More generally, Johannes' original post included the question "why does
this error occur"? 

I think the answer is two-fold: 

1. From Stata's point of view, the error is the user's, in trying to run
a program when the executable is not in sync. 

2. More positively, -xtivreg-'s programmers guessed that some users
would try to do that,  so that they wanted to explain gently to such
users what the problem was, rather than -xtivreg- crashing messily when
it tried to use something not available. 

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

Johannes Geyer

Thanks Martin for the hint.

I guess you are correct. Now I realize that it is really the printing
not 
the estimation that produces 
the error. Because the vector e(b) has all the estimated coefficients in

it of course. And all other 
estimated results are produced as well.

When I type 

********************************
qui xtivreg n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n), fd
********************************

I don't get an error message. 

Martin Weiss 
 
> Ok, so I am willing to bet that the problem is with "C:\Program Files
> (x86)\Stata11\ado\updates\_\_b_table.mata" which is supposed to print 
the
> coefficient table and checks whether your executable is at least the
one
> born 3/9/2010, see its line 847. Since yours is older, -xtivreg-
cannot
> continue...
 
Johannes Geyer
 
> Thanks Martin.
> 
> I get the following, when I 
> 
> *****************
> update, query
> (contacting http://www.stata.com)
> 
> Stata executable
>     folder:               R:\stata11\
>     name of file:         Stata.exe
>     currently installed:  21 Oct 2009
>     latest available:     20 Apr 2010
> 
> Ado-file updates
>     folder:               R:\stata11\ado\updates\
>     names of files:       (various)
>     currently installed:  20 Apr 2010
>     latest available:     20 Apr 2010
> 
> Utilities updates
>     folder:               R:\stata11\utilities
>     names of files:       (various)
>     currently installed:  11 Feb 2010
>     latest available:     11 Feb 2010
> 
> Recommendation
>     Type -update all-
> *****************
> 
> And that seems ok, except for the executable. I agree, updates are 
> important!!! 
 
Martin Weiss 
 
> > -xtivreg- has been subject to revision twice since the release of 
Stata 
> 11
> > <see http://www.stata.com/help.cgi?whatsnew>, so not -update-ing for
a 

> long
> > time is probably not a good idea anyway. 
> > 
> > 
> > What does
> > 
> > 
> > *************
> > update
> > *************
> > 
> > give you? 

Johannes Geyer
 
> > I work with Stata IC 11.0 (installed on a network drive) and an
error 
> > occurred which I do not understand.
> > 
> > I tried the following code:
> > 
> > 
> 
************************************************************************
****
> > ******
> > webuse abdata
> > xtivreg n l2.n l(0/1).w l(0/2).(k ys) yr1981-yr1984 (l.n = l3.n), fd
> > 
> 
************************************************************************
****
> > ******
> > 
> > and get 
> > 
> > 
> 
************************************************************************
****
> > ******
> > G2SLS random-effects IV regression      Number of obs   =       611
> > Group variable: id      Number of groups        =       140
> > 
> > R-sq:  within  = 0.7145 Obs per group: min      =       4
> > between = 0.9978        avg     =       4.4
> > overall = 0.9929        max     =       6
> > 
> >         Wald chi2(14)   =       82243.97
> > corr(u_i, X)       = 0 (assumed)        Prob > chi2     =
0.0000
> > 
> > your Stata exectuable is out of date
> > type -update executable- at the Stata   prompt
> > r(498);
> > 
> 
************************************************************************
****
> > ******
> > 
> > My conclusion is: update! But why does this error occur with such a 
> > standard estimation routine? 
> > I tried the same code with Stata IC 10.1 which looks even "more" out

of 
> > date and there I get results 
> > without any complain. 
> > 
> > The problem is, that we work with a many-user  network license and
for 

> an 
> > update of Stata all users need to close 
> > it which sometimes takes a long time. 

*
*   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/