OpenMx Help

Posted on
No user picture. cjvanlissa Joined: 04/10/2019

How to specify a random intercept cross-lagged panel model in OpenMx?

Dear readers,

I am trying to specify an RI-CLPM in OpenMx. However, if I constrain the variance of the observed indicators to zero, the model implied covariance matrix is not positive definite - but these variances are supposed to be zero, as they are partialized into a time-invariant component and a time-variant component. E.g., the variance x1 <-> x1 should be split up into RIx <-> RIx and cx1 <-> cx1.

Can anyone help me correctly specify this model in OpenMx?

Posted on
No user picture. ppl Joined: 01/12/2023

Maximum Likelihood for Cross-Lagged Panel Models with Fixed Effects - Allison et al. 2017

Hey all,

I just started recently to use OpenMx for academic reasons. I'm currently trying to reproduce the empirical example of the paper of Allison et al. 2017 in which they use a Maximum Likelihood Structural Equation Model (ML-SEM) for dynamic panel models.
The authors provide the code of the following software packages: R-lavaan, Stata, Mplus, SAS-Proc Calis. However, I'm using OpenMx in R since it seems to be more flexible for changes and adjustments than lavaan.

Posted on
Picture of user. rabil Joined: 01/14/2010

Residual Variance as a Function of a Latent Variable

Imagine you have a simple one-factor measurement error model - just one latent variable u with say 5 indicators. How would you make the residual variance of the indicators a function of u? For example, a linear function of u. Is this possible using OpenMx?

Posted on
Picture of user. rabil Joined: 01/14/2010

Constraint Not Honored

I have a simple two-factor measurement error model. There is a constraint on the variance of one of the latent factors that it be positive. It easily finds a solution using mxTryHard but the variance estimate is always negative. What is the point of using a constraint if it is not honored? The negative estimate then causes problems when I try to bootstrap confidence intervals. It complains about quantile missing values and NaNs and won't produce a result unless I remove the variance from the confidence interval list.

Posted on
No user picture. cjvanlissa Joined: 04/10/2019

MxModel returns class MxRAMModel instead of MxModel?

I'm calling MxModel via do.call, with a list of arguments. Previously, this consistently returned objects of class MxModel. Now (probably after the latest OpenMx update), it returns MxRAMModel, which is inconvenient because not all exported methods work with MxRAMModel.

I'm rolling back my installation of OpenMx now to check if it's due to the latest update, but that takes a while.. in the meantime - does anyone know why I'm getting MxRAMModels and how to prevent that?

Posted on
Picture of user. jmatosv Joined: 04/25/2015

Model identified at solution but not identified at start values

Openmx follows an iterative method when executing mxCheckIdentification function. Therefore, on first iteration start values are used to solve model equation; then, the iterative process continues until the best (optimal) solution is obtained.

If we run mxCheckIdentification(pathrun) and the model is locally identified at solution, but we run mxCheckIdentification(pathmod) and the model is not locally identified at start values, would that be a reason of concern when evaluating the model?

Posted on
No user picture. Sedat KANADLI Joined: 08/15/2022

Residual Covariance Matrix

Dear Mike,
A colleague and I studied the factor structure of a scale using the metaSEM package. The reviewers asked us for residual covariance matrice of structural model we construct. I couldn't see a command for this in the metaSEM package. There is vcov but this command gives sampling covariance matrices. Is there a way to see residual covariance matrices?
Sincerely

Posted on
No user picture. Roya Joined: 10/25/2021

Non-positive definit matrices

Hello Mike,

I am applying the MASEM method to 70 articles to find the direct and indirect effects of a variable (I have only three variables, one outcome, one exposure, and one mediator). I have 67 non-positive definite matrices. I understand that excluding all those matrices is missing lots of data. What should I do in this case? I did the analysis, ignoring the positive definite assumption, and I got the results but I don´t know how reliable are the results.
I´d be grateful if you please help me to find a solution for that.

Posted on
Picture of user. pehkawn Joined: 05/24/2020

Dealing with duplicated data in a multilevel model

I have two datasets of teacher and student data, each representing a level in a multilevel model. The teachers have a unique link to a student taught in one specific class, and teachers have unique response items for each class. On the other hand, each student have only a single entry and do not have unique data for each class. This leads to the issue that teachers may teach the same student in more than one class, which creates a duplicated entry of the student if the datasets are merged.