Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: hireg vs esta


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: hireg vs esta
Date   Fri, 18 May 2007 18:31:05 -0400

Megan--
As you will have seen from our posts, Richard likes -hireg- fine; I
don't love it.  I like -esta- better; -ssc install estout- to get it.

On a more substantive note, I assume from your variable names you are
estimating the impact of services, size, and organizational form on
the profit margins of US hospitals. Looking only at the impact of
adding organizational form as a set of dummies {public, forprofit}
might well obscure large and interesting differences across hospital
types (and across time, if that dimension is available to you) in the
measured impact of the other explanatory variables, i.e. I would opt
for a fully interacted model (and perhaps a panel model if the data
allow).

For example, in the table below, the effects of size are quite
different--for-profit and nonprofit hospitals make more per patient
with higher admissions (admtot) and less with more beds (bdtot),
whereas public hospitals have the opposite pattern, though neither
size metric may matter for public hospitals:

esta gov nfp fp, mti(public nonprof forprofit)
------------------------------------------------------------
                     (1)             (2)             (3)
                  public         nonprof       forprofit
------------------------------------------------------------
admtot        -0.00000684      0.00000334       0.0000140**
                 (-0.87)          (1.62)          (2.83)

bdtot           0.0000561       -0.000205*      -0.000509*
                  (0.16)         (-1.97)         (-2.04)

cclab               0.141          0.0118          0.0664
                  (1.34)          (0.44)          (1.56)

ohsrg             -0.0598          0.0105         -0.0291
                 (-0.54)          (0.38)         (-0.65)

psyem             -0.0571         0.00269         -0.0162
                 (-0.79)          (0.14)         (-0.51)

psyop            -0.00175        -0.00124         -0.0193
                 (-0.02)         (-0.06)         (-0.44)

traum              -0.137*         0.0204          0.0241
                 (-2.08)          (1.19)          (0.80)

brnbd             -0.0839         -0.0828*         0.0995
                 (-0.85)         (-2.16)          (0.76)

mri                 0.157*         0.0295          0.0418
                  (2.24)          (1.50)          (1.46)

pet                 0.130         0.00888          0.0240
                  (1.76)          (0.48)          (0.69)

spect              0.0497          0.0302         0.00875
                  (0.78)          (1.74)          (0.32)

_cons              -0.223***      -0.0743***      -0.0806**
                 (-4.12)         (-3.99)         (-3.16)
------------------------------------------------------------
N                     254            1364             348
------------------------------------------------------------
t statistics in parentheses
* p<0.05, ** p<0.01, *** p<0.001


On 5/18/07, Megan McHugh <mmchugh@gwu.edu> wrote:
"hireg dollarmargin (EDadmit surgical anyoutlier ortho cardiac surgcardiac surg ortho) (resbed DSH beds public forprofit levelone)"
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index