OpenMx General Help
jkarch
Joined: 03/15/2011
Product of Variables Identification
my aim is to use product of variables to fit a random-intercept random-slope model with latent mean centering. To be precise, my model is as follows. $i$ indexes person, $t$ indexes observations within a person.
$$Y_{it} = \beta_i (X_{it}-meanX_i) + meanY_i + residualY_{ij}$$
$$X_{it} = meanX_i + residualX_{ij} $$
$$\beta_i \sim N(\mu_\beta, \sigma_\beta^2), \quad meanX_i \sim N(\mu_X, \sigma_X^2), \quad meanY_i \sim N(\mu_Y, \sigma_Y^2)$$
- Read more about Product of Variables Identification
- 3 comments
- Log in or register to post comments
Manuel Rein
Joined: 04/01/2024
Sampling weights for mxFitFunctionMultigroup()
I was wondering if there is a way to specify sampling weights in a multi-group model? I want all observations in a certain group to be weighed equally, but groups with a larger weight should have more influence on the estimation of parameters that are shared/equal between groups.
Best,
Manuel
Paul Hoon
Joined: 02/17/2024
Seek copy of MX varimax rotation code posted years ago at VA Tech web site
Several years back we posted a varimax - quartimax script that worked well under an earlier implementation of MX.
1) Can you direct me to a site so I can retrieve this script and its documentation?
2) Would like to revise the script so it will run under R (Win 11) in the latest MX configuration. Could we get permission to get help from Neale and other administrators to guide us in making revisions to get the script to work in 2024?
Thank you for your advice and direction, best, Paul Hoon
Jaewonee
Joined: 10/02/2023
Can't solve dimname associated error
pheno2 <- read.csv("pheno2.csv")
library(OpenMx)
colnames(pheno2) <- gsub("\\.", "_", colnames(pheno2))
any(is.na(pheno2))
- Read more about Can't solve dimname associated error
- 9 comments
- Log in or register to post comments
lf-araujo
Joined: 11/25/2020
Can´t get the fit indices for DoC model, as Saturated DZ gets non pos def
lf-araujo
Joined: 11/25/2020
Can't derive CIs in multiple groups twin model. Too nested!
**Summary**: if I try go get cis for parameters in multiple groups twin models (written in matrix algebra only), open mx complains about it being too nested. Multiple groups here refers to running the model for males and females, for example. See MWE.
**Error**:
Lisdin
Joined: 03/08/2016
Unable to install OpenMx with NPSOL
I have been trying to install the OpenMx Team's build of OpenMx with the NPSOL optimizer, using the following line as per instructions (https://openmx.ssri.psu.edu/installing-openmx):
source('https://vipbg.vcu.edu/vipbg/OpenMx2/software/getOpenMx.R')
I have tried several times and with several R versions (4.1.3, 4.2.3, and 4.3.1), but I always get the warning that the package ‘OpenMx’ is not available as a binary package for this version of R (see full warning message attached).
- Read more about Unable to install OpenMx with NPSOL
- 1 comment
- Log in or register to post comments
Veronica_echo
Joined: 02/23/2018
Trouble with importing specific functions from OpenMx into my package
I am currently developing an R package called nlpsem, and I am trying to use some functions from the OpenMx package. Specifically, I am trying to use the mxOption and imxReportProgress functions. However, I am encountering some issues when trying to use these functions without explicitly loading the OpenMx package with library(OpenMx).
In my DESCRIPTION file, I have the following lines:
Imports:
OpenMx, ggplot2, dplyr, tidyr, stringr, Matrix, nnet, readr, methods
Ben
Joined: 06/20/2023
Error message when searching two words or phrase in OpenMx Forum
"Forbidden
You don't have permission to access /search/node/TERM1 TERM2 on this server."
Google doesn't seem to search the posts itself but only the forum overview.
lf-araujo
Joined: 11/25/2020
Announcing umxCLPM() TESTERS NEEDED
All I really want is another single-author paper complaining about cookie-cutter models, so I wrote the ultimate cookie-cutter (\s) for CLPM:
https://github.com/tbates/umx/commit/26a9cf01da4cc789be3fc66ddf16ab40050cf87d
It is hard to find spect details for some of these models on the original papers, so I need testers. In particular, I wonder whether the 'uv' correlation is really equated across in Heise's and in STARTS. I believe it is fixed for Hamaker's.
Enjoy.
- Read more about Announcing umxCLPM() TESTERS NEEDED
- Log in or register to post comments