Re: st: RE: Re: pwcompare

[Janet Hill <janethill73@yahoo.co.uk>]

Amir,

Thanks. I must go back to the theory as I thought I was asking the same question. The data that I had hoped to be working on will contain missing values hence my interest in xtmixed.

Martyn


----- Original Message -----
From: amir gahremanpour <gamirali@hotmail.com>
To: statalist <statalist@hsphsun2.harvard.edu>
Cc: 
Sent: Monday, 28 November 2011, 2:05
Subject: RE: st: RE: Re: pwcompare

Janet,

your anova command is different from xtmixed (you are not asking the same thing from xtmixed and anova)! If the groups are not balanced or you have missing data xtmixed is a better too, however the best person to ask is Phil Ender from UCLA group. I would email him directly.

regards

Amir



> Date: Sun, 27 Nov 2011 20:23:57 +0000
> From: janethill73@yahoo.co.uk
> Subject: Re: st: RE: Re: pwcompare
> To: statalist@hsphsun2.harvard.edu
> 
> Tony,
> 
> Thanks for that. I will go back to West, Welch & Galecki to try and improve my understanding of the underlying theory.
> 
> I am now totally confused as to how to compare the groups after using xtmixed in a repeated measures scenario, can anybody offer me any suggestions or advice?
> 
> Thank you,
> Janet
> 
> 
> ----- Original Message -----
> From: "Lachenbruch, Peter" <Peter.Lachenbruch@oregonstate.edu>
> To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
> Cc: 
> Sent: Saturday, 26 November 2011, 18:44
> Subject: st: RE: Re: pwcompare
> 
> I didn't see an answer to this.á I suspect it is related to the fact that the standard errors in the xtmixed model account for the random effects, and so are larger.á Note that the contrasts are the same in both as they are the sample means.
> 
> Tonyh
> 
> ________________________________________
> From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Janet Hill [janethill73@yahoo.co.uk]
> Sent: Friday, November 25, 2011 8:35 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Re: pwcompare
> 
> Why does pwcompare give different results after anova and xtmixed? Using the following syntax:
> xtmixed lhist i.group##i.time || dog:, reml cov(id)
> or
> anova lhist group / dog|group time time#group, repeated(time)
> followed by
> pwcompare group, emptycells(reweigh) eff asbalanced post
> 
> I get the following for xtmixed
> ------------------------------------------------------------------------------
> á á á á á á  |á á á á á á á á á á á á á á Unadjustedá á á á á  Unadjusted
> á á á á á á  |á  Contrastá  Std. Err.á á á zá á P>|z|á á  [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> lhistá á á á |
> á á á  group |
> á á  2 vs 1á |á -.4591643á  .5681373á á -0.81á  0.419á á -1.572693á á .6543643
> á á  3 vs 1á |á  1.200595á á .525993á á  2.28á  0.022á á  .1696677á á 2.231522
> á á  4 vs 1á |á -.3857097á á .525993á á -0.73á  0.463á á -1.416637á á .6452175
> á á  3 vs 2á |á  1.659759á  .5681373á á  2.92á  0.003á á  .5462306á á 2.773288
> á á  4 vs 2á |á  .0734546á  .5681373á á  0.13á  0.897á á -1.040074á á 1.186983
> á á  4 vs 3á |á -1.586305á á .525993á á -3.02á  0.003á á -2.617232á  -.5553775
> ------------------------------------------------------------------------------
> and for the anova
> ------------------------------------------------------------------------------
> á á á á á á  |á á á á á á á á á á á á á á Unadjustedá á á á á  Unadjusted
> á á á á á á  |á  Contrastá  Std. Err.á á á tá á P>|t|á á  [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> á á á  group |
> á á  2 vs 1á |á -.4591643á  .1047385á á -4.38á  0.000á á -.6722564á  -.2460723
> á á  3 vs 1á |á  1.200595á á .096969á á 12.38á  0.000á á á 1.00331á á  1.39788
> á á  4 vs 1á |á -.3857097á á .096969á á -3.98á  0.000á á -.5829947á  -.1884248
> á á  3 vs 2á |á  1.659759á  .1047385á á 15.85á  0.000á á  1.446667á á 1.872851
> á á  4 vs 2á |á  .0734546á  .1047385á á  0.70á  0.488á á -.1396375á á .2865467
> á á  4 vs 3á |á -1.586305á á .096969á  -16.36á  0.000á á  -1.78359á á -1.38902
> ------------------------------------------------------------------------------
> 
> I can see that the standard errors are different, but the question arises should I use xtmixed or anova for repeated measures?
> 
> I am using Stata 12.0, update 10 Nov 2011.
> 
> Thanks,
> Janet
> 
> *
> *á  For searches and help try:
> *á http://www.stata.com/help.cgi?search
> *á http://www.stata.com/support/statalist/faq
> *á http://www.ats.ucla.edu/stat/stata/
> *
> *á  For searches and help try:
> *á http://www.stata.com/help.cgi?search
> *á http://www.stata.com/support/statalist/faq
> *á 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/