Martin - Yes, sorry - I should have given the full anova command. Here is full output for a reduced data set that gives the same error:
Al
. list hr ag id time,noobs
+-------------------------+
| hr ag id time |
|-------------------------|
| 84.2 0 1392 0 |
| 90.8 0 1392 4 |
| 76.1 0 1392 -7 |
| 63.6 0 6291 0 |
| 62.3 0 6291 -7 |
|-------------------------|
| 66.4 0 6291 4 |
| 77.3 1 4011 -7 |
| 82.1 1 4011 0 |
| 77.6 1 4011 4 |
| 51.6 1 5695 0 |
|-------------------------|
| 51.5 1 5695 4 |
| 47.9 1 5695 -7 |
| 59.7 1 6867 -7 |
| 57.1 1 6867 0 |
| 49.5 1 6867 4 |
|-------------------------|
| 73.8 1 7108 4 |
| 64.5 1 7108 -7 |
| 76.4 1 7108 0 |
| 72.8 1 7218 0 |
| 74.5 1 7218 4 |
|-------------------------|
| 66.2 1 7218 -7 |
+-------------------------+
anova hr ag/id|ag time
Number of obs = 21 R-squared = 0.9268
Root MSE = 4.23546 Adj R-squared = 0.8780
Source | Partial SS df MS F Prob > F
-----------+----------------------------------------------------
Model | 2724.43049 8 340.553811 18.98 0.0000
|
ag | 302.399954 1 302.399954 0.65 0.4565
id|ag | 2323.83339 5 464.766678
-----------+----------------------------------------------------
time | 98.1971419 2 49.0985709 2.74 0.1049
|
Residual | 215.269643 12 17.9391369
-----------+----------------------------------------------------
Total | 2939.70013 20 146.985006
. test,showorder
Order of columns in the design matrix
1: _cons
2: (ag==0)
3: (ag==1)
4: (time==-7)
5: (time==0)
6: (time==4)
7: (id==1392)*(ag==0)
8: (id==4011)*(ag==1)
9: (id==5695)*(ag==1)
10: (id==6291)*(ag==0)
11: (id==6867)*(ag==1)
12: (id==7108)*(ag==1)
13: (id==7218)*(ag==1)
. matrix list cc
cc[2,13]
c1 c2 c3 c4 c5 c6 c7 c8 c9 c10 c11 c12 c13
r1 1 1 0 0 0 0 0 0 0 0 0 0 0
r1 1 0 1 0 0 0 0 0 0 0 0 0 0
. test,test(c) mtest
( 1) _cons + ag[1] = 0
---------------------------------------
| F(df,12) df p
-------+-------------------------------
(1) | 555.60 1 0.0000 #
-------+-------------------------------
all | 555.60 1 0.0000
---------------------------------------
# unadjusted p-values
. test,test(c) mtest matvlc(M)
option matvlc() not allowed
r(198);
. .
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Martin Weiss
Sent: Tuesday, February 24, 2009 11:46 AM
To: [email protected]
Subject: st: AW: matvlc option not allowed after -anova-
<>
Well, can you frame the problem in terms of an example that is accessible to all? It is hard to create one for oneself. Your -anova- is in replay mode, right?
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Feiveson, Alan H. (JSC-SK311)
Gesendet: Dienstag, 24. Februar 2009 18:18
An: [email protected]
Betreff: st: matvlc option not allowed after -anova-
Hi - I can't seem to get the syntax correct for the -matvlc- optin after anova. Here's what the help file says under anova postestimation:
Syntax for test after anova
test, test(matname) [mtest[(opt)] matvlc(matname)] syntax 1
matvlc(matname), a programmer's option, saves the variance-covariance matrix
of the linear combinations involved in the suite of tests. For the test
Lb = c, what is returned in matname is LVL', where V is the estimated
variance-covariance matrix of b.
But when I try the -matvlc- option, I get an error:
Can anyone tell me what's wrong?
Thanks
Al Feiveson
. anova
Number of obs = 45 R-squared =
0.8711
Root MSE = 5.62952 Adj R-squared =
0.7975
Source | Partial SS df MS F Prob >
F
-----------+----------------------------------------------------
Model | 5997.12926 16 374.820579 11.83
0.0000
|
ag | 1212.91789 1 1212.91789 38.27
0.0000
id|ag | 4084.81901 13 314.216847 9.91
0.0000
time | 699.39236 2 349.69618 11.03
0.0003
|
Residual | 887.361024 28 31.6914652
-----------+----------------------------------------------------
Total | 6884.49028 44 156.465688
. test,showorder
Order of columns in the design matrix
1: _cons
2: (ag==0)
3: (ag==1)
4: (time==-7)
5: (time==0)
6: (time==4)
7: (id==1392)*(ag==0)
8: (id==4011)*(ag==1)
9: (id==5695)*(ag==1)
10: (id==6291)*(ag==0)
11: (id==6867)*(ag==1)
12: (id==7108)*(ag==1)
13: (id==7218)*(ag==1)
14: (id==7330)*(ag==0)
15: (id==7694)*(ag==1)
16: (id==8138)*(ag==0)
17: (id==8168)*(ag==0)
18: (id==8920)*(ag==1)
19: (id==9098)*(ag==0)
20: (id==9203)*(ag==1)
21: (id==9997)*(ag==0)
. matrix list c
c[2,21]
c1 c2 c3 c4 c5 c6
c7 c8
r1 1 1 0 .33333333 .33333333 .33333333
.14285714 0
r1 1 0 1 .33333333 .33333333 .33333333
0 .125
c9 c10 c11 c12 c13 c14
c15 c16
r1 0 .14285714 0 0 0 .14285714
0 .14285714
r1 .125 0 .125 .125 .125 0
.125 0
c17 c18 c19 c20 c21
r1 .14285714 0 .14285714 0 .14285714
r1 0 .125 0 .125 0
. test,test(c) mtest
( 1) _cons + ag[1] + .3333333 time[1] + .3333333 time[2] + .3333333 time[3] + .1428571 id[1]*ag[1
> ] + .1428571 id[4]*ag[1] + .1428571 id[8]*ag[1] + .1428571
> id[10]*ag[1] +
.1428571 id[11]*ag[1] +
> .1428571 id[13]*ag[1] + .1428571 id[15]*ag[1] = 0
( 2) _cons + ag[2] + .3333333 time[1] + .3333333 time[2] + .3333333 time[3] + .125 id[2]*ag[2] +
> .125 id[3]*ag[2] + .125 id[5]*ag[2] + .125 id[6]*ag[2] + .125
> id[7]*ag[2]
+ .125 id[9]*ag[2] + .1
> 25 id[12]*ag[2] + .125 id[14]*ag[2] = 0
---------------------------------------
| F(df,28) df p
-------+-------------------------------
(1) | 3883.24 1 0.0000 #
(2) | 3313.40 1 0.0000 #
-------+-------------------------------
all | 3598.32 2 0.0000
---------------------------------------
# unadjusted p-values
. test,test(c) mtest matvlc(M)
option matvlc() not allowed
r(198);
.
*
* 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/
*
* 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/
*
* 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/
