Re: st: outreg2.versionSet error

[Nick Cox <njcoxstata@gmail.com>]

In addition to John Gallup's comments:

Michael's syntax 3b combines a call to -xi:- with factor variable
notation. StataCorp recommend against that.

The help for -outreg2- does specify the author's preferred way of
receiving error reports

    1. it's best if you send an example using -sysuse- or -webuse- data
    2. create a trace log file containing the error
    3. Send them to whoever wrote the thing and ask nicely.

and that's good advice.

Nick

On Mon, Mar 18, 2013 at 11:00 PM, Michael LeGower <mjl88@pitt.edu> wrote:
> There appears to be a flaw in the versionSet program within the outreg2
> package as installed by a -ssc install outreg2- command.  In the copy I
> have, any Stata version greater than 8.2 it treats as version 8.2, which
> it then passes back to the estimation command (if using the -outreg2
> <contents> : <estimation command>- syntax).  This seems to conflict with
> estimation commands that use factor variable notation.
>
> Here is an example using the 1978 Automobile reference dataset under Stata
> 12.1.
>
> which outreg2
> *c:\ado\plus\o\outreg2.ado
> *! outreg2 2.2.4 12apr2010 byroywada@hotmail.com
> *! based on outreg 3.0.6/4.0.0 byjohn_gallup@alum.swarthmore.edu
>
> sysuse auto
>
> * Syntax 1: Regression with factor variables; Fine in Stata 12, even when
> followed with an -outreg2- command separately
>
> regress price i.foreign
> outreg2 using test1
>
> * Syntax 2: -outreg2- prefix before same command; error results
>
> outreg2 using test2: regress price i.foreign
>
> * Syntax 3: -outreg2- prefix before same command with -xi- prefix; Fine
>
> outreg2 using test2: xi: regress price i.foreign
>
> * Syntax 3b: outreg2 prefix before command with -xi- prefix AND factor
> variable interaction; error
>
> outreg2 using test2: xi: regress price i.foreign#i.rep78
>
> With the stock -outreg2- command as installed from SSC, the second syntax
> fails and the first and third run fine.  Adding an interaction term makes
> even the third syntax fail, however.  If the SSC -outreg2- package could
> be updated so that commands 2 and 3b did not produce an error, that would
> be tremendous.  Otherwise, this is to let people know that an -xi- will be
> required if using the -outreg2- prefix before an estimation command with
> factor variable notation, and models with factor variable interactions
> will require you to run the -outreg2- command post estimation.
>
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/