Categorical Outcomes

Posted on
No user picture. Tom Rosenstrom Joined: 03/01/2018

Pairwise maximum likelihood estimator available?

Hi.
This is probably a good place to ask if anyone has implemented the new pairwise maximum likelihood estimator for Open Mx? For example, this paper mentions that "PLM fit estimates can also be obtained with Open Mx":
https://doi.org/10.3389/fpsyg.2016.00528

Posted on
Picture of user. Julia Joined: 03/29/2012

Estimating concordance

Hi.
As far as I understand, it is possible to estimate concordance of a dichotomous trait through a liability threshold model. If so, could anyone guide me a bit how to do that? I'm totally lost and don't know how to incorporate that into a saturated model.
I'm interested in pairwise and casewise concordance.

Best,
Julia

Posted on
No user picture. falkcarl Joined: 10/29/2015

lower bounds on slopes of dichotomous and graded item response models

When doing estimation of item factor analysis models in OpenMx, I'm wondering if there's a way to override the lower boundary on the item slopes (currently 1e-6, I think) for the dichotomous and graded item response models. e.g., imagine one wants to fit a multidimensional model, perhaps w/ crossloadings, or one has reverse worded items and it is expected that some item slopes will go negative. I believe these bounds may be set in the rpf package as rpf.ParamInfo returns this lower bound for the rpf.drm and rpf.grm item models.
Posted on
Picture of user. Rolandgem Joined: 10/24/2015

Univariate Binary Model with multiple covariates that are binary, categorical, and continuous.

This is my first attempt at binary univariate models with multiple covariates. When i fit the saturated model I get the following error which I am not sure what it actually means:

Error in runHelper(model, frontendStart, intervals, silent, suppressWarnings, :
MxComputeConfidenceInterval: reference fit is not finite

My guess is that it has something to do with the values or bounds?
What are some guidelines for choosing values when you have binary variables (as well as a mix of binary and continuous covariates)?

Posted on
Picture of user. eivind Joined: 11/04/2009

Categorical data with membership probabilities

Hello,
I have ordinal data where the probability for membership in each category is known (e.g. individual i has probability p for being member of category 1 of variable u). Does anybody know if it is possible to utilize this information in OpenMx, for example by weighting the thresholds? The probabilities would be different across twins and across thresholds of the ordinal.
Posted on
Picture of user. Natalja Joined: 02/26/2014

Errors in script twinAceOrd.R

Hello everyone,

I am trying to learn how to fit a univariate ACE model when the variable of interest is ordinal. For this I've started with the script called Univariate Twin Ordinal-Matrix/twinAceOrd.R that I found under the TC 2012 - OpenMx website. When I run the code i get the following four error messages:

1st,
> # Generate Descriptive Statistics
> #colMeans(mzDataOrd,na.rm=TRUE)
> #colMeans(dzDataOrd,na.rm=TRUE)
> cov(mzDataOrd,use="complete")
Error: is.numeric(x) || is.logical(x) is not TRUE
> cov(dzDataOrd,use="complete")
Error: is.numeric(x) || is.logical(x) is not TRUE

Posted on
Picture of user. iloo Joined: 05/26/2010

Definition variables for categorical and continuous variables, something's wrong

I've noticed this a while ago but forgot to post it to you developers.

Something goes wrong when you include definition variables in a model where both continuous and categorical variables are included, and use the definition variables for regression on the mean. For example running this code yields on my machine:

library(OpenMx)
N <- 2000
u <- rbinom(N,1,.5)
x <- .5*u+rnorm(N)
y <- mxFactor( rbinom(N,1,pnorm(-2+u)) , levels=c(0,1) )

model <- mxModel( 'BinCont',
mxMatrix('Full',nrow=1,ncol=2,free=c(T,T),name='Betas'),

Posted on
No user picture. lingsuer87 Joined: 05/03/2013

ERROR for missing value where TRUE/FALSE needed

Dear all.

I am trying to add age as covariant into my liability threshold model. the script runs well before I add age covariant, However, after I add age covariate into the model, OpenMx always give me this error message:
Error in if (label %in% fixedVars && startVals[[label]] != value) { :
missing value where TRUE/FALSE needed.

Posted on
No user picture. Lisa Joined: 06/07/2013

Proband ascertainment

Dear all

I have a twin dataset with proband ascertainment (i.e. at least one proband in each pair is affected). I cant seem to figure out how in incorporate this in my ACE model - can anyone help? I have found several examples involving ACE models in twin data (for example http://openmx.psyc.virginia.edu/svn/tags/stable-1.2/demo/UnivariateTwinAnalysis_PathRaw.R) however none of these speak of how to take proband ascertainment into account.

Posted on
No user picture. lingsuer87 Joined: 05/03/2013

If I have more than 3 levels for my ordinal variables then I have N-1 increments?

Hi! guys,

I am doing the threshold model for ordinal variables. and I finished the variables for 3 levels. However, I need modify the script to another variable which contain more 5 levels. I sucess with saturated model wiithout increments It gives me 18 parameters : rMZ,rDZ,4 threshold for MZ1 4 threshold for MZ2,4 threshold for DZ1 and 4 threshold for DZ2,respectively. But I was stuck at ADE model. Some problems in increments.

My script like this:

# Set Starting Values

thVals <-c(-0.226,-0.099,0.637,2.628) # start value for thresholds t1,t2,t3,t4.