General SEM Discussions
Ryne
Joined: 07/31/2009
Help with Feature Request: Saturated Model Helper Functions
Hi all,
Based on a discussion in another thread (http://openmx.psyc.virginia.edu/thread/505), I'm soliciting feature request details for a helper function that builds saturated and independence models.
leylay
Joined: 06/03/2011
alex76
Joined: 05/26/2011
model comparison mediator model
Hello,
I have a question on the following: I have a model with 6 IV's, 3 mediators M, and 1 DV. (all observed variables).
Suppose model 1 has all direct effects of the IV's on the M's, and all direct effects of the M's on the DV. All IV's are allowed to correlate.
Model 2 is the same, except that one IV is dropped from the model.
How should I compare these two models to each other? If I understand this correctly, the models are not nested? So should I then use the AIC instead of a Chi sqr difference?
Kind regards, Alexandra
- Read more about model comparison mediator model
- 3 comments
- Log in or register to post comments
myaseen208
Joined: 07/21/2010
Stochastic three-mode models for mean and covariance structures
Dear All Members,
I'm trying to learn Stochastic three-mode models methodology proposed by Oort (1999)
http://onlinelibrary.wiley.com/doi/10.1348/000711099159099/abstract http://www.google.com/search?aq=f&sourceid=chrome&ie=UTF-8&q=Stochastic+Three-Mode+Models+for+Mean+and+Covariance+Structures#sclient=psy&hl=en&source=hp&q=%22Stochastic+Three-Mode+Models+for+Mean+and+Covariance+Structures%22&aq=f&aqi=&aql=&oq=&pbx=1&bav=on.2,or.r_gc.r_pw.&fp=2ec23705adf4e555.
manu
Joined: 10/08/2009
sample weights
Dear all,
I am working with data from a study with a complex sample design. This requires the use of sample weights (which are known). Is this possible in OpenMx?
In a post by neale from 09/24/10 it has been suggested to extend the mxFIMLObjective() function so that sample weights, which are part of the dataframe, are available as an argument whithin the function. I think that is exactly what I am looking for, but I could not find any further information. Can you point me to any?
- Read more about sample weights
- 15 comments
- Log in or register to post comments
tbates
Joined: 07/31/2009
non-normal SEM (third moment)
Hi,
Did anyone see this paper:
Koken Ozaki, Hideki Toyoda, Norikazu Iwama, Saori Kubo and Juko Ando (2011). Using Non-Normal SEM to Resolve the ACDE Model in the Classical Twin Design
http://www.springerlink.com/content/24480h65724464t7/
Thoughts? They argue that estimating skewness (3rd moment of a distribution) in addition to mean and variance, the extra degree of freedom makes ACDE models identified...
This seems over value: Can we do it in OpenMx?
t
- Read more about non-normal SEM (third moment)
- 2 comments
- Log in or register to post comments
Jean
Joined: 11/10/2009
Estimation method
Hi,
Can we choose estimation method when conducting SEM, such as OLS, GMM besides ML, FIML?
FIML assumes errors have a multivariate normal distribution. If the errors are not normally distributed, the FIML method might produce poor results, while 2SLS and 3SLS do not assume a specific distribution for the errors. Do we have a option of estimation method in OpenMx?
- Read more about Estimation method
- 1 comment
- Log in or register to post comments
hr.punio
Joined: 01/20/2011
Non-recursive model and negative SMC
Is this OK? SMC is negative. SMC for reduced form is positive.
Reference to the literature will be appreciated
--t
- Read more about Non-recursive model and negative SMC
- 3 comments
- Log in or register to post comments
jone1087
Joined: 11/05/2010
Mx Status Red error for a regression model
Hello,
I am running two different regression models. The first runs fine, but the second returns the Mx Status Red error. The Model 2 script is the same as Model 1, just with different data. Any help would be appreciated.
Thanks,
Jeff
# begin script
> ###########
> # Model 1 #
> ###########
>
>
> R.smp1 <- matrix(c(1.0000000, 0.7534779, 0.3378677,
+ 0.7534779, 1.0000000, 0.3757937,
+ 0.3378677, 0.3757937, 1.0000000),3,3)
>
> pred <- c("x1","x2"); out <- "y"
> varnames <- c(pred,out)
- Read more about Mx Status Red error for a regression model
- 1 comment
- Log in or register to post comments
paologhisletta
Joined: 01/19/2010
Software to draw SEM diagrams
Greetings,
it appears that the OpenMx team prefers the Graphviz software to draw SEM diagrams. I guess the choice was a compromise between quality and flexibility (Graphviz appears to run on several OS).
What drawing software would you suggest under Windows XP if I wanted to create a diagram from scratch? (I looked at Tetrad, which seems less esthetic than Graphviz). Is there anything as good as OmniGraffle for Macs? Or would you suggest installing a Mac emulator on Windows and use OmniGraffle? Please don't suggest I go Linux or Mac ;-)
- Read more about Software to draw SEM diagrams
- 10 comments
- Log in or register to post comments