|
|
|
||||
|
|
|||||
Note: The mvtest command is a rather old user written program that does not work with current version of Stata. We recommend the use of the manova command which was added to Stata beginning with version 9.
This web page was modified on 9/13/2007 to incorporate the use of manova
Here is an example of multivariate regression tests using both mvreg and manova with the hsb2 dataset.
use http://www.ats.ucla.edu/stat/stata/notes/hsb2
mvreg read write socst = math science female
Equation Obs Parms RMSE "R-sq" F P
----------------------------------------------------------------------
read 200 4 7.205288 0.5136 68.98202 0.0000
write 200 4 6.532664 0.5322 74.31583 0.0000
socst 200 4 8.858612 0.3294 32.0913 0.0000
------------------------------------------------------------------------------
| Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
read |
math | .4808247 .0704142 6.83 0.000 .3419581 .6196914
science | .3662612 .0671489 5.45 0.000 .233834 .4986884
female | .1024223 1.033879 0.10 0.921 -1.936533 2.141377
_cons | 7.870521 3.223374 2.44 0.016 1.513573 14.22747
-------------+----------------------------------------------------------------
write |
math | .4058757 .0638409 6.36 0.000 .2799725 .531779
science | .3422707 .0608805 5.62 0.000 .2222058 .4623357
female | 5.960556 .9373646 6.36 0.000 4.11194 7.809171
_cons | 10.41243 2.922467 3.56 0.000 4.648914 16.17595
-------------+----------------------------------------------------------------
socst |
math | .4688891 .0865714 5.42 0.000 .2981581 .6396201
science | .2372759 .0825569 2.87 0.004 .074462 .4000898
female | 1.985517 1.271112 1.56 0.120 -.5212957 4.49233
_cons | 14.33547 3.963008 3.62 0.000 6.519861 22.15108
------------------------------------------------------------------------------
manova read write socst = math science female, continuous(math science female)
Number of obs = 200
W = Wilks' lambda L = Lawley-Hotelling trace
P = Pillai's trace R = Roy's largest root
Source | Statistic df F(df1, df2) = F Prob>F
-----------+--------------------------------------------------
Model | W 0.3263 3 9.0 472.3 30.67 0.0000 a
| P 0.7540 9.0 588.0 21.93 0.0000 a
| L 1.8201 9.0 578.0 38.96 0.0000 a
| R 1.6756 3.0 196.0 109.47 0.0000 u
|--------------------------------------------------
Residual | 196
-----------+--------------------------------------------------
math | W 0.7307 1 3.0 194.0 23.83 0.0000 e
| P 0.2693 3.0 194.0 23.83 0.0000 e
| L 0.3685 3.0 194.0 23.83 0.0000 e
| R 0.3685 3.0 194.0 23.83 0.0000 e
|--------------------------------------------------
science | W 0.7999 1 3.0 194.0 16.18 0.0000 e
| P 0.2001 3.0 194.0 16.18 0.0000 e
| L 0.2502 3.0 194.0 16.18 0.0000 e
| R 0.2502 3.0 194.0 16.18 0.0000 e
|--------------------------------------------------
female | W 0.8199 1 3.0 194.0 14.20 0.0000 e
| P 0.1801 3.0 194.0 14.20 0.0000 e
| L 0.2196 3.0 194.0 14.20 0.0000 e
| R 0.2196 3.0 194.0 14.20 0.0000 e
|--------------------------------------------------
Residual | 196
-----------+--------------------------------------------------
Total | 199
--------------------------------------------------------------
e = exact, a = approximate, u = upper bound on F
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services