OpenMx Developer Forums
pdeboeck
Joined: 08/04/2009
Crashing R in OS X
Here's an interesting one...
So I've recently been working on converting an mx script to OpenMx (file attached). I might be doing something I shouldn't, but the script caused R on my Mac to close w/out explanation or warning. Everything goes fine until I hit "mxRun." I ran this under Windows XP and got the following error:
> EstModel <- mxRun(CoupledModel)
Running CoupledModel
Error in mxRun(CoupledModel) :
BLAS/LAPACK routine 'DGEMM ' gave error code -13
- Read more about Crashing R in OS X
- 6 comments
- Log in or register to post comments
mspiegel
Joined: 07/31/2009
loading packages conditionally
The OpenMx repository has a work-in-progress feature that generates graphs from RAM style models. This package requires the Rgraphviz library. I don't want to add Rgraphviz as a required library. Not all users will need to make graphs, and it turns out that the Rgraphviz library can be a pain to install on Windows (you need a specific version of the Graphviz library that is hidden on their website). So this feature conditionally works if the Rgraphviz package is installed.
- Read more about loading packages conditionally
- 4 comments
- Log in or register to post comments
pdeboeck
Joined: 08/04/2009
Bootstrapping Confidence Intervals
I couldn't find info on bootstrapping confidence intervals on the site, so I wanted to put in a plug for such an option. Kris Preacher and I were talking about this yesterday.
- Read more about Bootstrapping Confidence Intervals
- 2 comments
- Log in or register to post comments
tbates
Joined: 07/31/2009
UnivariateTwinAnalysis_MatrixRaw.R
There are close to zero comments on this script, and much of it is quite obscure.
1. There are still places with T instead of TRUE and unlabeled parameters relying on being in the right place:
mxMatrix("Full", 1, 2, T, 20, "mean", dimnames=list(NULL, selVars), name="expMeanMZ"),
2. Most matrices need a comment about what it is they are holding, i.e.:
mxMatrix("Full", nrow=1, ncol=1, free=TRUE, values=.6, label="a", name="X") # what is this for?
mxAlgebra(X %*% t(X), name="A"), # what does this do?
Proposing an algebra for the the groups is critical and needs a comment:
- Read more about UnivariateTwinAnalysis_MatrixRaw.R
- 3 comments
- Log in or register to post comments
tbrick
Joined: 07/31/2009
Ordinal Data Proposal (#2)
I've been talking to Ryne about this, and I think we've put together a fairly reasonable proposal for the user experience for ordinal data.
We welcome comments and suggestions.
A) From the pathy perspective:
- Read more about Ordinal Data Proposal (#2)
- 2 comments
- Log in or register to post comments
tbates
Joined: 07/31/2009
mxPath
topic for discussing mxPath() function
- Read more about mxPath
- 23 comments
- Log in or register to post comments
Steve
Joined: 07/30/2009
So many AICs, so little time. What's a summary() to do?
There has been an ongoing discussion on both the optimizer forum and the developer's mailing list about which AIC to use. This thread is here to gather the discussion in one place.
Since the major SEM packages differ in the way they calculate AIC, I proposed we print something like:
AIC (Mx) = -123456 AIC (Amos) = 654321 AIC (MPlus) = 234567 AIC (LISREL) = 132435
Let the AIC discussions begin!
tbates
Joined: 07/31/2009
using mxModel
questions about mxModel
- Read more about using mxModel
- 18 comments
- Log in or register to post comments
tbates
Joined: 07/31/2009
hangovers from Mx
There are places in the demo scripts that inherit things from Mx that are not necessary: like the fact that Mx couldn't read files with a labels line..
So in
data(twinData)
twinVars <- c('fam','age','zyg','part','wt1','wt2','ht1','ht2','htwt1','htwt2','bmi1','bmi2')
Much easier to say
data(twinData) twinVars = names(twinData) # maybe with a comment so people know what is going on # 'fam','age','zyg','part','wt1','wt2','ht1','ht2','htwt1','htwt2','bmi1','bmi2'
- Read more about hangovers from Mx
- 1 comment
- Log in or register to post comments
tbates
Joined: 07/31/2009
Underlying model for diagrams: Multiple group etc.
I guess the GUI will have to create a fairly universal underlying script model, allowing it to create the full-spectrum of models.
What underlying script model is the GUI expecting to generate/read? (i.e., should we think about a RAM+ that can handle all models, or pick a suite that covers the bases, expanding as the GUI capability expands?