Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

 From Paul Burkander To statalist@hsphsun2.harvard.edu Subject Re: st: R: variable not found? Date Mon, 24 Oct 2011 16:53:26 -0400

```The use of -test- in this case is not unnecessary, and if it's
incorrect I'd rather you tell me why.  Please see
http://www.ats.ucla.edu/stat/stata/dae/mvreg.htm for an explanation of
the usefulness of the joint test across equations.  This exercise
follows the example on that page almost exactly.

Also please note that this question has nothing to do with the
substance of the homework, but with what still appears to be a bug in
stata. I should also note that the homework has been handed in, so no
advice given here will affect it.

Exactly what I did: using emacs, I replaced the spaces with commas.
then in stata:

insheet using filename
manova motive aspire algebra geom =group
mvreg
test group

What exactly did you type that didn't result in error?

Thank you.
Paul

. insheet using data.csv
(5 vars, 221 obs)

. des

Contains data
obs:           221
vars:             5
size:         4,641 (99.9% of memory free)
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
storage  display     value
variable name   type   format      label      variable label
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
motiv           float  %9.0g                  MOTIV
aspire          float  %9.0g                  ASPIRE
algebra         float  %9.0g                  ALGEBRA
geom            float  %9.0g                  GEOM
group           byte   %8.0g                  GROUP
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Sorted by:
Note:  dataset has changed since last saved

. manova motiv aspire algebra geom = group

Number of obs =     221

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
-----------+--------------------------------------------------
group | W   0.9056      1     4.0   216.0     5.63 0.0002 e
| P   0.0944            4.0   216.0     5.63 0.0002 e
| L   0.1043            4.0   216.0     5.63 0.0002 e
| R   0.1043            4.0   216.0     5.63 0.0002 e
|--------------------------------------------------
Residual |               219
-----------+--------------------------------------------------
Total |               220
--------------------------------------------------------------
e = exact, a = approximate, u = upper bound on F

. mvreg

Equation          Obs  Parms        RMSE    "R-sq"          F        P
----------------------------------------------------------------------
motiv             221      2    17.81392    0.0384   8.740918   0.0035
aspire            221      2    14.48455    0.0920   22.19203   0.0000
algebra           221      2    17.94691    0.0318    7.20056   0.0078
geom              221      2    14.22695    0.0615   14.34351   0.0002

------------------------------------------------------------------------------
|      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
motiv        |
1.group |   7.413426   2.507497     2.96   0.003     2.471512    12.35534
_cons |   42.30231    2.01703    20.97   0.000     38.32703    46.27758
-------------+----------------------------------------------------------------
aspire       |
1.group |   9.604709   2.038852     4.71   0.000     5.586426    13.62299
_cons |   62.79179   1.640052    38.29   0.000     59.55949     66.0241
-------------+----------------------------------------------------------------
algebra      |
1.group |   6.778811   2.526216     2.68   0.008     1.800006    11.75762
_cons |   45.65538   2.032087    22.47   0.000     41.65043    49.66033
-------------+----------------------------------------------------------------
geom         |
1.group |   7.584382   2.002592     3.79   0.000     3.637563     11.5312
_cons |   66.57667   1.610884    41.33   0.000     63.40185    69.75149
------------------------------------------------------------------------------

. test group
r(111);

On Mon, Oct 24, 2011 at 2:16 PM, Steven Samuels <sjsamuels@gmail.com> wrote:
>
> Sorry, my reply was incomplete.  I get no error message with -test- -mvreg- after pasting your data into BBEdit, replacing commas with spaces, and making all names lower case. You haven't actually shown us, as the FAQ request, exactly what you typed.
>
>
> Steve
>
>
> Paul,
>
>
>
> Your use of the -test- statement appears to be incorrect, and unnecessary too.  The output  of -manova- do give the (four) tests that all groups have zero coefficiernts
>
> You must test four coefficients, not one, if you want a global test of group differences.
> I'd also remind the group that the FAQ suggest we don't respond to requests for help with homework.
>
>
> Steve
>
>
> On Oct 24, 2011, at 4:14 PM, Paul Burkander wrote:
>
> In my attempt at brevity, I didn't note that the variables are turned
> to lowercase when pasted, and we were running the command with the
> correct case.  And the error is actually "regressor group not found"
> My apologies for the erroneous description.
>
> This seems like a bug.  It's replicated in stata 12.
>
> On Mon, Oct 24, 2011 at 6:01 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
>> Whatever the problem is, that will not solve it.
>>
>> -format- changes display format only; it doesn't otherwise change the status of any variable, especially its kind, type or storage characteristics.
>>
>> Nick
>> n.j.cox@durham.ac.uk
>>
>> Christian Otchia
>>
>> You should format the variable after pasting in Stata 11.
>>
>> Try this:
>>
>> format %6.1g GROUP
>>
>> test GROUP
>>
>>
>> Paul Burkander
>>
>>> I'm trying to help my wife do some stata homework, and we've run into
>>> some odd behavior.
>>>
>>> Below is a sample of the data.  She's supposed to do a manova of
>>> MOTIV,ASPIRE,ALGEBRA,GEOM, on GROUP , then mvreg, then test the
>>> restriction that the coefficients on GROUP are zero in all four
>>> equations.  After doing the manova and mvreg, we typed test GROUP, but
>>>
>>> Any ideas?  Perhaps we're doing something dumb?  She entered the data
>>> by pasting into the data editor.  We're using Stata 11 SE and I just
>>> did a full update.
>>>
>>> Thanks, Paul
>>>
>>> MOTIV,ASPIRE,ALGEBRA,GEOM,GROUP
>>> 74.31,78.57,68.59,78.57,1
>>> 27.15,61.91,37.15,71.43,1
>>> 64.31,90.48,67.16,83.34,1
>>> 40.01,64.29,48.59,59.53,1
>>> 35.73,59.53,31.44,54.76,1
>>> 55.73,83.34,60.02,85.72,0
>>> 52.87,64.29,62.88,69.05,0
>>> 58.59,71.43,64.31,80.95,0
>>
>> *
>> *   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/
>

*
*   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/
```