Dear Statalist,
I have found that when I run an -mlogit- model and specify a base category other than the default, the results do not display properly when I use esttab.
I run the following code:
use http://www.stata-press.com/data/r11/sysdsn1, clear
xi i.nonwhite i.site
eststo clear
eststo: mlogit insure _Inonwhite_1 _Isite_2 _Isite_3 if male, base(1)
eststo: mlogit insure _Inonwhite_1 _Isite_2 _Isite_3 if !male, base(2)
esttab, unstack
esttab, unstack drop(o.*)
But my output from the second esttab (with the option drop(o.*)) is missing a column. There should be a fourth column entitled "Indemnity":
(1) (2)
insure insure
Prepaid Uninsure Uninsure
------------------------------------------------------------
_Inonwhite_1 1.024 -0.307 -0.615
(1.87) (-0.27) (-1.33)
_Isite_2 0.310 -2.111 -0.891
(0.76) (-1.92) (-1.62)
_Isite_3 -0.680 -1.019 0.690
(-1.50) (-1.37) (1.52)
_cons 0.224 -0.826* -1.657***
(0.76) (-2.04) (-4.66)
------------------------------------------------------------
N 154 462
Does anyone know why this may be occurring? I have run across in a number of different specifications using different data sets. The issue seems to arise from the way Stata stores the coefficients for the omitted base categories, but I can't figure out why.
Thanks
Jamie Fogel
*
* 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/
