Hello everyone,
consider the following
which produces the following output (I'm skipping the non-important commands)
There are two binary explanatory variables foreign and heavy, and thus four groups
Thank you.
consider the following
Code:
sysuse auto, clear gen byte heavy = weight > 3000 regress price mpg i.foreign##i.heavy lincom 1.foreign + 1.foreign#1.heavy gen byte nonhvyfor = (!heavy & foreign) gen byte hvydom = (heavy & !foreign) gen byte hvyfor = (heavy & foreign) regress price mpg nonhvyfor hvydom hvyfor lincom hvyfor - hvydom
Code:
. regress price mpg i.foreign##i.heavy
Source | SS df MS Number of obs = 74
-------------+------------------------------ F( 4, 69) = 8.91
Model | 216345984 4 54086495.9 Prob > F = 0.0000
Residual | 418719412 69 6068397.28 R-squared = 0.3407
-------------+------------------------------ Adj R-squared = 0.3024
Total | 635065396 73 8699525.97 Root MSE = 2463.4
-------------------------------------------------------------------------------
price | Coef. Std. Err. t P>|t| [95% Conf. Interval]
--------------+----------------------------------------------------------------
mpg | -211.8713 68.7862 -3.08 0.003 -349.096 -74.6466
|
foreign |
Foreign | 1710.16 842.3177 2.03 0.046 29.78266 3390.538
1.heavy | 1074.217 911.9885 1.18 0.243 -745.1499 2893.585
|
foreign#heavy |
Foreign#1 | 3483.295 1985.39 1.75 0.084 -477.4488 7444.039
|
_cons | 9508.832 1842.119 5.16 0.000 5833.906 13183.76
-------------------------------------------------------------------------------
. lincom 1.foreign + 1.foreign#1.heavy
( 1) 1.foreign + 1.foreign#1.heavy = 0
------------------------------------------------------------------------------
price | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 5193.456 1794.605 2.89 0.005 1613.317 8773.594
------------------------------------------------------------------------------
. regress price mpg nonhvyfor hvydom hvyfor
Source | SS df MS Number of obs = 74
-------------+------------------------------ F( 4, 69) = 8.91
Model | 216345984 4 54086495.9 Prob > F = 0.0000
Residual | 418719412 69 6068397.28 R-squared = 0.3407
-------------+------------------------------ Adj R-squared = 0.3024
Total | 635065396 73 8699525.97 Root MSE = 2463.4
------------------------------------------------------------------------------
price | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
mpg | -211.8713 68.7862 -3.08 0.003 -349.096 -74.6466
nonhvyfor | 1710.16 842.3177 2.03 0.046 29.78266 3390.538
hvydom | 1074.217 911.9885 1.18 0.243 -745.1499 2893.585
hvyfor | 6267.673 1969.226 3.18 0.002 2339.175 10196.17
_cons | 9508.832 1842.119 5.16 0.000 5833.906 13183.76
------------------------------------------------------------------------------
. lincom hvyfor - hvydom
( 1) - hvydom + hvyfor = 0
------------------------------------------------------------------------------
price | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
(1) | 5193.456 1794.605 2.89 0.005 1613.317 8773.594
------------------------------------------------------------------------------
- nonhvdom those cars that are domestic and not heavy (base group)
- hvydom those cars that are domestic and heavy
- nonhvfor those cars that are foreign and not heavy
- hvyfor those cars that are foreign and heavy
Thank you.