Re: st: outreg adds minus symbol to se's

[David Harless <dwharles@vcu.edu>]

Wouter van Gils wrote:
> Hi,
>
> I'm a newby in using Stata, but like it very much. A few days ago I 
> tried the outreg package to export my results to Excel. The export of 
> coefficients works fine, but the standard errors turn negative when I 
> use the 'coefastr' option. (The coefastr tag marks the coefficients with 
> significance levels instead of the se's or t-values.)
>
> For instance:
>
> regress ZMINCULT VAGE VEDUC ZHHINC
> outreg using C:\temp\zmincult2.xls, bdec(2) se nolabel replace
>
> results in:
>
> VAGE    0.03
>     (0.00)**
> VEDUC    0.11
>     (0.01)**
> ZHHINC    0.18
>     (0.02)**
> Const    -2.36
>     (0.10)**
>
> while,
>
> regress ZMINCULT VAGE VEDUC ZHHINC
> outreg using C:\temp\zmincult2.xls, bdec(2) se nolabel coefastr replace
>
> results in:
>
> VAGE    0.03**
>     0.00
> VEDUC    0.11**
>     -0.01
> ZHHINC    0.18**
>     -0.02
> Const    -2.36**
>     -0.1
>
>
> The SE for VEDUC and the Constant turned negative. The same applies when 
> I use t-values below the coefficients. This does not happen when I 
> choose a regular *.out file or an *.doc output, it only happens when I 
> choose *.xls export. Is this a bug in the outreg package? Can anyone 
> recreate this error on another machine?
>
> Thx in advance,
>
> Wouter van Gils
Wouter,
This is an MS Excel problem, not an outreg problem. Just try pasting (0.10) into 
Excel; it takes (0.10) as representing a negative number.  There may be some way 
of formatting the cell in Excel to make it come out as you wish, but I haven't 
found it.
Dave Harless
*
*   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/