Dynamic Factor Analysis

[E. Bernardo]

Any spss macro or syntax that will do dynamic factor analysis?

I've been trying to google but turned unsuccessful.

Comments are welcome.

===================== 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

[Art Kendall]

Since different disciplines have different dialects of statistics, I
suggest that you briefly describe the procedure you would like to do, the circumstances in which you have see it used, etc.

Also please describe the data and questions you would like to use it for.

[David Marso]

http://www.bing.com/search?q=dynamic+factor+analysis&pc=MOZI&form=MOZSBR
None that I know of.  Within this search query I note discussion of a few R packages.
Good Luck.  I suspect someone with deep skills could devise an SPSS Matrix solution.
----

E. Bernardo wrote

Any spss macro or syntax that will do dynamic factor analysis?I've been trying to google but turned unsuccessful.Comments are welcome.

=====================
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

[Bruce Weaver]

It's no wonder some of us have never heard of it--according to one of the sources in David's search results (http://www.stata.com/meeting/2italian/Federici.pdf), it is not well known in the "Anglo-Saxon literature".  ;-)

David Marso wrote

http://www.bing.com/search?q=dynamic+factor+analysis&pc=MOZI&form=MOZSBR
None that I know of.  Within this search query I note discussion of a few R packages.
Good Luck.  I suspect someone with deep skills could devise an SPSS Matrix solution.
----

E. Bernardo wrote

Any spss macro or syntax that will do dynamic factor analysis?I've been trying to google but turned unsuccessful.Comments are welcome.

=====================
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

[David Marso]

Seems also that it is more likely to be utilized in Econometrics than say Psychology or Sociology.
Wonder how this substantively differs from a longitudinal SEM?
---

Bruce Weaver wrote

It's no wonder some of us have never heard of it--according to one of the sources in David's search results (http://www.stata.com/meeting/2italian/Federici.pdf), it is not well known in the "Anglo-Saxon literature".  ;-)

David Marso wrote

http://www.bing.com/search?q=dynamic+factor+analysis&pc=MOZI&form=MOZSBR
None that I know of.  Within this search query I note discussion of a few R packages.
Good Luck.  I suspect someone with deep skills could devise an SPSS Matrix solution.
----

E. Bernardo wrote

Any spss macro or syntax that will do dynamic factor analysis?I've been trying to google but turned unsuccessful.Comments are welcome.

=====================
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

[Art Kendall]

But it could have many many uses in the social and behavioral sciences.
Spatial and temporal auto-correlation take into account  when a variable is measured repeatedly across places and /  or across time points.  I do not see why it could not also be used to across other forms of repeats at points on a continuum, such as auditory frequencies, the electromagnetic spectrum etc.

These could be used in many areas outside economics such as policy analysis, attitude drift, program evaluation, crime monitoring, climate modeling, etc. etc.

[Mark Miller]

This is a jargon problem.  Such issues have been studied under various guises

since at least Anderson and Rubin in 1949. Applications are widespread.

Look for Latent Class Growth Models,   trajectory models, reduced-rank regression

Current examples

      -- group-based trajectory model by Nagin ( SAS/Stata Proc TRAJ) 

     -- growth mixture model by Muthen (Mplus)

... Mark

On Fri, Dec 12, 2014 at 9:46 AM, Art Kendall <[hidden email]> wrote:

But it *could *have many many uses in the social and behavioral sciences.
Spatial and temporal auto-correlation take into account  when a variable is
measured repeatedly across places and /  or across time points.  I do not
see why it could not also be used to across other forms of repeats at points
on a continuum, such as auditory frequencies, the electromagnetic spectrum
etc.

These could be used in many areas outside economics such as policy analysis,
attitude drift, program evaluation, crime monitoring, climate modeling, etc.
etc.



-----
Art Kendall
Social Research Consultants
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Dynamic-Factor-Analysis-tp5728157p5728168.html

Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
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

[E. Bernardo]

Art Kendal and all,

I have three time series continuous variables (X, Y and Z); monthly data for three years.  In theory, these three variables are indicators of a single unknown factor/latent variable. So in this exercise, I will conduct a confirmatory factor analysis for my three time series variables. Because the variables are time series, some people suggested me to use a Dynamic Factor Analysis, which I think is not available in the point-click of SPSS. My goal of conducting the analysis is to determine the factor loadings of X,Y and Z on the factor F. Some people call it Dynamic Factor Analysis.

On Saturday, December 13, 2014 2:36 AM, Mark Miller <[hidden email]> wrote:

This is a jargon problem.  Such issues have been studied under various guises

since at least Anderson and Rubin in 1949. Applications are widespread.

Look for Latent Class Growth Models,   trajectory models, reduced-rank regression

Current examples

      -- group-based trajectory model by Nagin ( SAS/Stata Proc TRAJ) 

     -- growth mixture model by Muthen (Mplus)

... Mark

On Fri, Dec 12, 2014 at 9:46 AM, Art Kendall <[hidden email]> wrote:

But it *could *have many many uses in the social and behavioral sciences.
Spatial and temporal auto-correlation take into account  when a variable is
measured repeatedly across places and /  or across time points.  I do not
see why it could not also be used to across other forms of repeats at points
on a continuum, such as auditory frequencies, the electromagnetic spectrum
etc.

These could be used in many areas outside economics such as policy analysis,
attitude drift, program evaluation, crime monitoring, climate modeling, etc.
etc.



-----
Art Kendall
Social Research Consultants
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Dynamic-Factor-Analysis-tp5728157p5728168.html

Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
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

===================== 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