Repeated measures in mixed

[Maguin, Eugene]

I have an odd question. A repeated measures design can be analyzed in Mixed
using the Repeated subcommand. The documentation has an example.

My question is this. Can a repeated measures design be analyzed using Mixed
but WITHOUT the Repeated subcommand such that the same results as an
analysis using the Repeated subcommand? I'm assuming the answer is yes and
my followup questions are what is the command syntax and what is the
underlying alebraic model.

Gene Maguin

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Roy Money]

I suppose it depends on how you  want to analyze the data.  I generally
use the REPEATED statement because as I understand it that is the way to
get estimated marginal means for the successive times, which are
presumably fixed.  This kind of analysis is known as a covariance
pattern model analysis where you specify a type of correlation for the
within person observations in the REPEATED statement.

If the times are not fixed or you do not need the marginal means for
comparisons then the RANDOM statement allows for you to model random
intercepts and slopes for each individual and assess how much between
person variance there is  in the intercepts and slopes.  I understand
you can even combine the two approaches.  But either approach allows you
to use all available data with less restrictive assumptions about the
within person variance.

There is a great discussion of these different approaches with their
associated algebra  in
http://gseweb.harvard.edu/~faculty/singer/Papers/sasprocmixed.pdf

See also Singer and Willett, Applied Longtitudinal Analysis for a more
thorough book length discussion of these issues.

Roy Money


Gene Maguin wrote:

> I have an odd question. A repeated measures design can be analyzed in Mixed
> using the Repeated subcommand. The documentation has an example.
>
> My question is this. Can a repeated measures design be analyzed using Mixed
> but WITHOUT the Repeated subcommand such that the same results as an
> analysis using the Repeated subcommand? I'm assuming the answer is yes and
> my followup questions are what is the command syntax and what is the
> underlying alebraic model.
>
> Gene Maguin
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
>

--
Roy Money  M.S.
Programmer Analyst
Department of Psychiatry
Yale University School of Medicine
The Consultation Center
389 Whitney Avenue
New Haven, CT 06511
ph:[phone] x 126
fax: [phone]

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[SR Millis-3]

Although it is possible to extend the random coefficient model by adding a correlation structure to the model, it should be used with caution.  There is the danger of "over-correction." For example, it is possible to model both a random intercept and an additional exchangeable correlation structure.  Because these 2 options are exactly the same, this will lead to over-correction and the associated problems with the estimation and interpretation of the regression coefficients.

~~~~~~~~~~~
Scott R Millis, PhD, ABPP (CN,CL,RP), CStat, CSci
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Dept of Emergency Medicine
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email:  [hidden email]
Email:  [hidden email]
Tel: [phone]
Fax: [phone]


--- On Wed, 12/2/09, Roy Money <[hidden email]> wrote:

> From: Roy Money <[hidden email]>
> Subject: Re: Repeated measures in mixed
> To: [hidden email]
> Date: Wednesday, December 2, 2009, 4:38 PM
> I suppose it depends on how you
> want to analyze the data.  I generally
> use the REPEATED statement because as I understand it that
> is the way to
> get estimated marginal means for the successive times,
> which are
> presumably fixed.  This kind of analysis is known as a
> covariance
> pattern model analysis where you specify a type of
> correlation for the
> within person observations in the REPEATED statement.
>
> If the times are not fixed or you do not need the marginal
> means for
> comparisons then the RANDOM statement allows for you to
> model random
> intercepts and slopes for each individual and assess how
> much between
> person variance there is  in the intercepts and
> slopes.  I understand
> you can even combine the two approaches.  But either
> approach allows you
> to use all available data with less restrictive assumptions
> about the
> within person variance.
>
> There is a great discussion of these different approaches
> with their
> associated algebra  in
> http://gseweb.harvard.edu/~faculty/singer/Papers/sasprocmixed.pdf
>
> See also Singer and Willett, Applied Longtitudinal Analysis
> for a more
> thorough book length discussion of these issues.
>
> Roy Money
>
>
> Gene Maguin wrote:
> > I have an odd question. A repeated measures design can
> be analyzed in Mixed
> > using the Repeated subcommand. The documentation has
> an example.
> >
> > My question is this. Can a repeated measures design be
> analyzed using Mixed
> > but WITHOUT the Repeated subcommand such that the same
> results as an
> > analysis using the Repeated subcommand? I'm assuming
> the answer is yes and
> > my followup questions are what is the command syntax
> and what is the
> > underlying alebraic model.
> >
> > Gene Maguin
> >
> > =====================
> > To manage your subscription to SPSSX-L, send a message
> to
> > [hidden email]
> (not to SPSSX-L), with no body text except the
> > command. To leave the list, send the command
> > SIGNOFF SPSSX-L
> > For a list of commands to manage subscriptions, send
> the command
> > INFO REFCARD
> >
> >
>
> --
> Roy Money  M.S.
> Programmer Analyst
> Department of Psychiatry
> Yale University School of Medicine
> The Consultation Center
> 389 Whitney Avenue
> New Haven, CT 06511
> ph:[phone] x 126
> fax: [phone]
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email]
> (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the
> command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Ryan]

A random intercept model is mathematically equivalent to a model with a repeated statement with a
specified compound symmetric variance-covariance matrix. Consequently, the following two pieces of
code will yield identical results:

MIXED y BY time
  /FIXED=time | SSTYPE(3)
  /METHOD=REML
  /RANDOM=INTERCEPT | SUBJECT(person) COVTYPE(VC).

MIXED y BY time
  /FIXED=time | SSTYPE(3)
  /METHOD=REML
  /REPEATED=time | SUBJECT(person) COVTYPE(CS).

Ryan

Gene Maguin wrote

I have an odd question. A repeated measures design can be analyzed in Mixed
using the Repeated subcommand. The documentation has an example.

My question is this. Can a repeated measures design be analyzed using Mixed
but WITHOUT the Repeated subcommand such that the same results as an
analysis using the Repeated subcommand? I'm assuming the answer is yes and
my followup questions are what is the command syntax and what is the
underlying alebraic model.

Gene Maguin

=====================
To manage your subscription to SPSSX-L, send a message to
[e-mail] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD