RE: st: Anova and pwcompare and not estimable

 From Garry Anderson To "statalist@hsphsun2.harvard.edu" Subject RE: st: Anova and pwcompare and not estimable Date Mon, 9 Jan 2012 22:04:36 +0000

```Dear Statalist and Martyn,

Thank you for referring me to the excellent discussion.

I have used -pwcompare calib#shape, asbalanced emptycells(reweight)- and obtained the following.

I would have expected two different standard errors for the two types of comparison (within calib, and between calib). The -anova- defines which error term to use for each factor and I would have thought this carries over to comparisons of means.

set more of
use http://www.stata-press.com/data/r12/t77
gen sub1to6 = subject
replace sub1to6=sub1to6 + 3 if calib==2
anova score calib / sub1to6|calib shape calib#shape,repeated(shape)

pwcompare calib#shape, asbalanced emptycells(reweight)

Pairwise comparisons of marginal linear predictions

Margins      : asbalanced
Empty cells  : reweight

-----------------------------------------------------------------
|   Contrast   Std. Err.     [95% Conf. Interval]
----------------+------------------------------------------------
calib#shape |
(1 2) vs (1 1)  |         -1   .9077853     -2.977894    .9778942
(1 3) vs (1 1)  |          3   .9077853      1.022106    4.977894
(1 4) vs (1 1)  |   .6666667   .9077853     -1.311227    2.644561
(2 1) vs (1 1)  |          3   .9077853      1.022106    4.977894
(2 2) vs (1 1)  |   1.333333   .9077853     -.6445608    3.311227
(2 3) vs (1 1)  |   4.666667   .9077853      2.688773    6.644561
(2 4) vs (1 1)  |   5.333333   .9077853      3.355439    7.311227
(1 3) vs (1 2)  |          4   .9077853      2.022106    5.977894
(1 4) vs (1 2)  |   1.666667   .9077853     -.3112275    3.644561
(2 1) vs (1 2)  |          4   .9077853      2.022106    5.977894
(2 2) vs (1 2)  |   2.333333   .9077853      .3554392    4.311227
(2 3) vs (1 2)  |   5.666667   .9077853      3.688773    7.644561
(2 4) vs (1 2)  |   6.333333   .9077853      4.355439    8.311227
(1 4) vs (1 3)  |  -2.333333   .9077853     -4.311227   -.3554392
(2 1) vs (1 3)  |   2.66e-15   .9077853     -1.977894    1.977894
(2 2) vs (1 3)  |  -1.666667   .9077853     -3.644561    .3112275
(2 3) vs (1 3)  |   1.666667   .9077853     -.3112275    3.644561
(2 4) vs (1 3)  |   2.333333   .9077853      .3554392    4.311227
(2 1) vs (1 4)  |   2.333333   .9077853      .3554392    4.311227
(2 2) vs (1 4)  |   .6666667   .9077853     -1.311227    2.644561
(2 3) vs (1 4)  |          4   .9077853      2.022106    5.977894
(2 4) vs (1 4)  |   4.666667   .9077853      2.688773    6.644561
(2 2) vs (2 1)  |  -1.666667   .9077853     -3.644561    .3112275
(2 3) vs (2 1)  |   1.666667   .9077853     -.3112275    3.644561
(2 4) vs (2 1)  |   2.333333   .9077853      .3554392    4.311227
(2 3) vs (2 2)  |   3.333333   .9077853      1.355439    5.311227
(2 4) vs (2 2)  |          4   .9077853      2.022106    5.977894
(2 4) vs (2 3)  |   .6666667   .9077853     -1.311227    2.644561
-----------------------------------------------------------------

.

Kind regards, Garry

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Martyn Sherriff
Sent: Monday, 9 January 2012 10:08 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: Anova and pwcompare and not estimable

pwcompare calib#shape, asbalanced emptycells(reweight)

There is an excellent discussion of this at http://www.ats.ucla.edu/stat/stata/faq/margins_nested1.htm

Cheers,
Martyn

On 9 January 2012 08:52, Garry Anderson <g.anderson@unimelb.edu.au> wrote:
> pwcompare calib#shape
*
*   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/
```