I don't know what exactly David means by "a simple split plot design".
However, the example posted beneath my signature below demonstrates the
use of -parmest- and -multproc- to show, using the Bonferroni procedure,
that non-US cars are lighter (on average) than US cars, at least in the
rep78 categories 3 and 4.
Obviously, the example below is not a one-liner. However, the software
would have to be psychic (not just clever) to know what parameters
represented "interesting" comparisons. And, as Nick observed, there
seems to be no consensus regarding the "correct" choice of a
multiple-test procedure, even among statisticians. However, -multproc-
give the user a choice of several procedures, and users can specify
their own choice of which parameters represent "interesting"
comparisons.
I hope this helps.
Roger
Roger B Newson
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
UNITED KINGDOM
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Email: [email protected]
Web page: www.imperial.ac.uk/nhli/r.newson/
Departmental Web page:
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/pop
genetics/reph/
Opinions expressed are those of the author, not of the institution.
**** BEGINNING OF Stata LOG - CUT HERE
. sysuse auto, clear
(1978 Automobile Data)
. keep if inlist(rep78,3,4,5)
(15 observations deleted)
. xi, prefix(_J) noomit i.rep78|foreign
i.rep78|foreign _JrepXforei_# (coded as above)
. regress weight _J*, noconst
Source | SS df MS Number of obs =
59
-------------+------------------------------ F( 6, 53) =
263.82
Model | 547865622 6 91310937 Prob > F =
0.0000
Residual | 18343577.8 53 346105.241 R-squared =
0.9676
-------------+------------------------------ Adj R-squared =
0.9639
Total | 566209200 59 9596766.1 Root MSE =
588.31
------------------------------------------------------------------------
------
weight | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
_Jrep78_3 | 3442.222 113.2198 30.40 0.000 3215.132
3669.312
_Jrep78_4 | 3532.222 196.1024 18.01 0.000 3138.891
3925.554
_Jrep78_5 | 1960 415.9959 4.71 0.000 1125.618
2794.382
_JrepXfore~3 | -1432.222 358.0323 -4.00 0.000 -2150.344
-714.1004
_JrepXfore~4 | -1324.444 277.3306 -4.78 0.000 -1880.699
-768.1897
_JrepXfore~5 | 443.3333 459.9008 0.96 0.339 -479.1109
1365.778
------------------------------------------------------------------------
------
. parmest, norestore label list(,)
+-----------------------------------------------------------------------
-----------------------------------------------+
| parm label estimate stderr dof
t p min95 max95 |
|-----------------------------------------------------------------------
-----------------------------------------------|
1. | _Jrep78_3 rep78==3 3442.2222 113.21975 53
30.403018 3.210e-35 3215.1322 3669.3123 |
2. | _Jrep78_4 rep78==4 3532.2222 196.10237 53
18.012135 2.994e-24 3138.8907 3925.5538 |
3. | _Jrep78_5 rep78==5 1960 415.99594 53
4.7115845 .00001816 1125.6178 2794.3822 |
4. | _JrepXforei_3 (rep78==3)*foreign -1432.2222 358.0323 53
-4.0002598 .00019743 -2150.3441 -714.10037 |
5. | _JrepXforei_4 (rep78==4)*foreign -1324.4444 277.33063 53
-4.7756877 .00001455 -1880.6992 -768.18965 |
|-----------------------------------------------------------------------
-----------------------------------------------|
6. | _JrepXforei_5 (rep78==5)*foreign 443.33333 459.90081 53
.96397597 .33943587 -479.11088 1365.7775 |
+-----------------------------------------------------------------------
-----------------------------------------------+
. multproc if strpos(label,"*foreign")>0, reject(rejbonf)
Method: bonferroni
Uncorrected overall critical P-value: .05
Number of P-values: 3
Corrected overall critical P-value: .01666667
Number of rejected P-values: 2
. list
+-----------------------------------------------------------------------
---------------------------------------------------------+
| parm label estimate stderr dof
t p min95 max95 rejbonf |
|-----------------------------------------------------------------------
---------------------------------------------------------|
1. | _Jrep78_3 rep78==3 3442.2222 113.21975 53
30.403018 3.210e-35 3215.1322 3669.3123 . |
2. | _Jrep78_4 rep78==4 3532.2222 196.10237 53
18.012135 2.994e-24 3138.8907 3925.5538 . |
3. | _Jrep78_5 rep78==5 1960 415.99594 53
4.7115845 .00001816 1125.6178 2794.3822 . |
4. | _JrepXforei_3 (rep78==3)*foreign -1432.2222 358.0323 53
-4.0002598 .00019743 -2150.3441 -714.10037 1 |
5. | _JrepXforei_4 (rep78==4)*foreign -1324.4444 277.33063 53
-4.7756877 .00001455 -1880.6992 -768.18965 1 |
|-----------------------------------------------------------------------
---------------------------------------------------------|
6. | _JrepXforei_5 (rep78==5)*foreign 443.33333 459.90081 53
.96397597 .33943587 -479.11088 1365.7775 0 |
+-----------------------------------------------------------------------
---------------------------------------------------------+
**** END OF Stata LOG - CUT HERE
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of David Airey
Sent: 11 March 2008 14:07
To: [email protected]
Subject: Re: st: RE: Stata has disappointing support of
contrasts/multiple comparisons in mixed ANOVA
.
Can you post such an analysis of a simple split plot design?
-Dave
On Mar 11, 2008, at 8:39 AM, Newson, Roger B wrote:
> I personally do a lot of work with multiple comparisons. However, I
> use
> -parmest-, -dsconcat- and -metaparm- to produce a dataset with 1
> observation per comparison, and -smileplot- and -multproc- to do the
> various multiple-test procedures.
>
> The -parmest- package (which includes -metaparm-), the -smileplot-
> package (wiich includes -multproc-), and the -dsconcat- package, can
> all
> be downloaded from SSC. More papers on all of these packages can be
> downloaded frommmy website (see my signature below).
>
> I hope this helps.
>
> Roger
>
>
> Roger B Newson
> Lecturer in Medical Statistics
> Respiratory Epidemiology and Public Health Group
> National Heart and Lung Institute
> Imperial College London
> Royal Brompton campus
> Room 33, Emmanuel Kaye Building
> 1B Manresa Road
> London SW3 6LR
> UNITED KINGDOM
> Tel: +44 (0)20 7352 8121 ext 3381
> Fax: +44 (0)20 7351 8322
> Email: [email protected]
> Web page: www.imperial.ac.uk/nhli/r.newson/
> Departmental Web page:
>
http://www1.imperial.ac.uk/medicine/about/divisions/nhli/respiration/pop
> genetics/reph/
>
> Opinions expressed are those of the author, not of the institution.
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Feiveson,
> Alan H. (JSC-SK311)
> Sent: 11 March 2008 13:26
> To: [email protected]
> Subject: st: Stata has disappointing support of contrasts/multiple
> comparisons in mixed ANOVA
>
> Hi - I have been repeatedly asked by people who I convinced to buy
> Stata how to do multiple comparisons of interaction effects under more
> "complicated" anova models such as those with at least one factor
> random
> and possibly with repeated measures, etc. Apparently Stata does not
> have
> a built-in command/option for multiple comparisons for anything but a
> one-way model and does not even support approximate methods for such
> comparisons other than using xtmixed and creating all the contrasts
> and
> tests "by hand". Even then, further processing has to be undertaken to
> account for multiple testing.
>
> Embarassingly enough, I have had to refer people to SAS to do these
> types of comparisons without a lot of customized programing. Am I
> missing something or is Stata really woefully deficient in this area?
>
> Al Feiveson
>
> *
> * 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/
>
> *
> * 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/
--
David C. Airey, Ph.D.
Pharmacology Research Assistant Professor
Center for Human Genetics Research Member
Department of Pharmacology
School of Medicine
Vanderbilt University
Rm 8158A Bldg MR3
465 21st Avenue South
Nashville, TN 37232-8548
TEL (615) 936-1510
FAX (615) 936-3747
EMAIL [email protected]
URL http://people.vanderbilt.edu/~david.c.airey/dca_cv.pdf
URL http://www.vanderbilt.edu/pharmacology
*
* 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/
*
* 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/
