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]

st: Interaction and constraint


From   Shige Song <shigesong@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Interaction and constraint
Date   Sun, 2 Jan 2011 14:48:30 -0500

Dear All,

I have a simple question regarding interaction terms and constraints.
Using data from the software manual, I did a simple regression model:

. use http://www.stata-press.com/data/r11/fvex
. regress y i.group group#c.age

      Source |       SS       df       MS              Number of obs =    3000
-------------+------------------------------           F(  5,  2994) =   21.15
       Model |  47475.1592     5  9495.03183           Prob > F      =  0.0000
    Residual |  1343957.85  2994  448.883716           R-squared     =  0.0341
-------------+------------------------------           Adj R-squared =  0.0325
       Total |  1391433.01  2999  463.965657           Root MSE      =  21.187

------------------------------------------------------------------------------
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
       group |
          2  |  -13.19784   3.880448    -3.40   0.001    -20.80646   -5.589225
          3  |  -11.14087   4.189539    -2.66   0.008    -19.35553   -2.926199
             |
 group#c.age |
          1  |  -.4751707   .0630779    -7.53   0.000    -.5988511   -.3514903
          2  |  -.2267166   .0624212    -3.63   0.000    -.3491095   -.1043237
          3  |  -.1848601   .0910187    -2.03   0.042    -.3633257   -.0063946
             |
       _cons |   90.55752   3.009782    30.09   0.000     84.65606    96.45897
------------------------------------------------------------------------------

Now I want to simply the model by constrainning the second
(2.group#c.age) and the third terms (3.group#c.age) to be equal
(suppose that will improve the model fit), how should I do this?

In addition, assume this is a longitudinal data analysis situation and
I am estimating a population average model using "xtlogit, pa", can I
constrain the two terms to be equal?

Many thanks.

Best,
Shige
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


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