OpenSEM Forums

I thought I might post this to see if anyone has come across a similar problem. In running a growth mixture model within OpenMx, I have been stymied by status code 10 despite trying mxTryHard (and mxTryHardWideSearch) and moving the starting values around by hand. The strange thing is that this occurs with actual data and with simulated data excluding missingness. Does anyone have a recommendation regarding starting values for a growth mixture model?

Hi,

I am running a trivariate ACE Cholesky model where trait 1 and 2 are positively correlated, but trait 3 is negatively correlated with 1 and 2. The problem is that when I run the model, the upperbound CI's for the phenotypic correlation due to a, c and e is over 1 for some estimates (e.g. MZM.h2[1,3] ubound= 1.14034072)

I have 3 other similar analyses with different traits, with similar correlations and the same problem.

Any ideas how to fix this?

Thank you very much!

# -------------------------------------------------------------------------------------------------

Hello,

The ACE model matrix spec documentation appears to be incorrect. Minor issue but want to bring to attention:
http://openmx.psyc.virginia.edu/docs/OpenMx/2.3.1/GeneticEpi_Matrix.html#ace-model-a-twin-analysis

The specified model is an ade model (.25%x%C in DZs) but the text refers to ace.

Hi OpenMX Community!

I have 3 traits, and I would like to assess whether maybe 2 come from the same genes, while the third has different genes.
That is, I would like to run a IP model with 2 Ac factors, one loading onto 2 traits, and the other one loading onto the last trait. I understand that this would not be identified.
-Is it possible to simplify this by having only 1 Ac factor that loads onto 2 traits, while the last trait only has As factors?
-Would I have to constrain the factor loadings of Ac to be equal since Ac only loads onto two traits?

Recently I was trying a trivariate Cholesky model with the latest version of OpenMx. However, when I estimated the standardized CI of parameters, it reported “Error in runHelper(model, frontendStart, intervals, silent, suppressWarnings, : NLOPT fatal error -1”. Then I followed the advices of Rob replied to another topic that change the optimizer from “SLSQP” to “NPSOL”. Although it run smoothly, the results contained minus value and bigger than 1 that were very strange, such as:

CholACE.StandardizedA[1,1] 4.835575e-02 0.4772359455 8.043944e-01

I have a question, which I closely related to an earlier question of me (http://openmx.psyc.virginia.edu/thread/3991). Given a longitudinal SEM model, which has been fitted using many persons, is there a recommended approach to obtained person-specific predictions for unobserved time-points, as there is for obtaining factor scores? (see e.g http://www.tandfonline.com/doi/abs/10.1080/00273171.2012.730072 by Ryne Estabrooka & Michael Neale).

Dear all,

I am very new to both OpenMx and structural equation modelling. I am trying to analyse data from educational research (see attached file for description). I have tried some examples from OpenMx but couldn’t figure out how to run SEM on my own data. I would appreciate greatly if I can get help on how to run factor SEM on my data. I have 3 factors (class, subject and student) and 12 measurements that are binomial (0 or 1). Thanks in advance

konval

New poster here, and new-ish user to twin analyses. I am trying to run an ACE model that I believe is called "scalar sex limited" -- I want to allow the total variance to differ across sex, but I want to constrain the A, C, and E to be the same so that I am reporting just one heritability estimate. Here are the key parts of my syntax:

twinACEModel <- mxModel("twinACE",
mxModel("ACE",
# ace path coefficients for males and females
mxMatrix("Lower", nrow=nv, ncol=nv, free=TRUE, values=0.2, label="am11", name="am"),