Adrian de la Garza wrote:
> I am running this model
>
> mlogit indic g_usip1 lswap embi_sd, basecat(0)
>
> and then using
>
> outreg using mlogit.out, replace bdec(3) bra nocon onecol
>
> to put my results in a table.
>
> My dependent variable takes values 0,1,2 with 0 as the base category of
> the mlogit. When I run the model, I see all the results, but my outreg
> command only stores those results for my 1s but not for my 2s.
>
> Why isn't -outreg- displaying the whole set of results?
For my part, all I can say is that, using the 'auto' dataset, I get the same:
. mlogit rep78 price mpg weight length foreign, basecat(1)
[...]
. outreg using junk.out, bdec(3) bra nocon onecol
Repair Record 1978
2:Price 0.001
[0.67]
2:Mileage (mpg) -0.054
[0.14]
2:Weight (lbs.) -0.004
[0.87]
2:Length (in.) 0.107
[0.89]
2:Car type -17.880
[0.00]
Observations 69
Absolute value of z statistics in brackets
* significant at 5%; ** significant at 1%
It appears to take the results from the first panel and omits the rest,
which is strange behaviour indeed!
Perhaps -est table- offers a partial way out here. Again, using 'auto':
. est table rep78, b(%9.3f) stats(r2_p ll chi2 N) star label varwidth(20)
newpanel
-------------------------------------
Variable | rep78
---------------------+---------------
1 |
Price | -0.000
Mileage (mpg) | 0.020
Weight (lbs.) | 0.001
Length (in.) | -0.010
Car type | -33.048
Constant | -0.392
---------------------+---------------
2 |
Price | 0.000
Mileage (mpg) | -0.034
Weight (lbs.) | -0.003
Length (in.) | 0.097
Car type | -33.928
Constant | -9.452
---------------------+---------------
4 |
Price | -0.000
Mileage (mpg) | 0.233
Weight (lbs.) | -0.001
Length (in.) | 0.085
Car type | 3.404*
Constant | -19.996
---------------------+---------------
5 |
Price | 0.000
Mileage (mpg) | 0.415*
Weight (lbs.) | -0.004
Length (in.) | 0.167
Car type | 3.667*
Constant | -33.950*
-------------------------------------
-------------------------------------
Statistics | rep78
---------------------+---------------
r2_p | 0.241
ll | -71.103
chi2 | 45.177
N | 69.000
-------------------------------------
legend: * p<0.05; ** p<0.01; *** p<0.001
Of course, this doesn't really answer your question about -outreg-,
especially since you probably want your regressions side-to-side, as
opposed to being stacked. Others may know of more effective workarounds
for your problem.
CLIVE NICHOLAS |t: 0(044)191 222 5969
Politics |e: [email protected]
Newcastle University |http://www.ncl.ac.uk/geps
*
* 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/
