I am getting somewhat different model results from an anova procedure and it's accompanying regression output. When regressing Y on x1 and x2, the significance levels reported for x1 and x2 are very different for anova vs. regression. My guess is that the procedures use somewhat different algorithms. Why are the results different and how should I interpret them? The output follows. Thanks in advance.
Linda Salisbury
University of Michigan
=============
. anova Y x1 x2 x1*x2
Number of obs = 99 R-squared = 0.0992
Root MSE = .688542 Adj R-squared = 0.0708
Source | Partial SS df MS F Prob > F
---------+----------------------------------------------------
Model | 4.96141459 3 1.65380486 3.49 0.0187
|
x1 | 2.8461798 1 2.8461798 6.00 0.0161
x2 | .302792702 1 .302792702 0.64 0.4262
x1*x2 | 1.93603708 1 1.93603708 4.08 0.0461
|
Residual | 45.0385854 95 .474090373
-----------+----------------------------------------------------
Total | 50.00 98 .510204082
. reg
Source | SS df MS Number of obs = 99
-------------+------------------------------ F( 3, 95) = 3.49
Model | 4.96141459 3 1.65380486 Prob > F = 0.0187
Residual | 45.0385854 95 .474090373 R-squared = 0.0992
-------------+------------------------------ Adj R-squared = 0.0708
Total | 50.00 98 .510204082 Root MSE = .68854
-------------------------------------------------------------------------
Y Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------------------------------------------------------------------
_cons 2.347826 .143571 16.35 0.000 2.062802 2.632851
x1
1 .0595813 .1953754 0.30 0.761 -.3282878 .4474504
2 (dropped)
x2
6 -.3913043 .20304 -1.93 0.057 -.7943897 .011781
12 (dropped)
x1*x2
1 6 .56082 .2775219 2.02 0.046 .0098694 1.111771
1 12 (dropped)
2 6 (dropped)
2 12 (dropped)
-------------------------------------------------------------------------
*
* 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/
