Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Regression with only higher level DV and IVs


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Regression with only higher level DV and IVs
Date   Sat, 5 Jan 2013 12:23:33 +0000

You might get away with using -egen- to summarize by school and then
-egen-'s -tag()- function to select precisely one observation for each
school.

Here is a silly example to show technique:

. sysuse auto
(1978 Automobile Data)

. egen wt_mean = mean(weight), by(rep78)

. egen mpg_mean = mean(mpg), by(rep78)

. egen tag = tag(rep78)

. regress mpg_mean wt_mean if tag

      Source |       SS       df       MS              Number of obs =       5
-------------+------------------------------           F(  1,     3) =  102.35
       Model |  43.0722735     1  43.0722735           Prob > F      =  0.0021
    Residual |  1.26244615     3  .420815385           R-squared     =  0.9715
-------------+------------------------------           Adj R-squared =  0.9620
       Total |  44.3347197     4  11.0836799           Root MSE      =   .6487

------------------------------------------------------------------------------
    mpg_mean |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     wt_mean |  -.0078468   .0007756   -10.12   0.002    -.0103151   -.0053785
       _cons |    45.1725   2.336426    19.33   0.000     37.73695    52.60805
------------------------------------------------------------------------------

But getting an appropriate structure need not be as painful as you
imply: see -collapse-, -contract-.

Nick

On Sat, Jan 5, 2013 at 11:59 AM, Stephen Cranney <scranney@sas.upenn.edu> wrote:

> I have a dataset with my observations being individual students who
> are then nested within schools. I am trying to do a regression of
> school-level dependent variables on school-level indicators.  Is there
> a quick and easy way to do this without resorting to painful data
> shaping or tricking the multilevel modelling commands into doing it?
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index