I see CIs now jitter if they can't get started. That's nice.

Often multivariate scripts won't get started when start values are specified with one value (because matrix is not invertable). I wrote a helper that jiggles starts to avoid this, but it would be nice if mx did this itself, if it can't get started.

This would also take away a lot of frustration from users whose scripts won't run, when the only problem is that the start values are identical across one or more matrices.

the old gen_epi helper tableFitStatistics(fit) would accept just one parameter, and describe that model's fit:
> tableFitStatistics(fit)
Name ep -2LL df AIC
Model 1 : top 7 2964.37 468 2028.37

The built-in replacement throws an error when the comparison list is omitted:
> mxCompare(fit)
Error in mxCompare(fit) :
'comparison' argument be a MxModel object or list of MxModel objects
It would be nice if it defaulted to simply give the base model data when no comparison models are provided...

Quote:

In order to implement alternative FIML objective functions, it would be nice to have an equivalent to R's QR decomposition.

summary() called on a model with an mxRowObjective dies with the message:

> summary(rmodRun)
Error in inherits(x, "factor") :
no slot of name "dims" for this object of class "MxRowObjective"


This is not a 1.0 priority, but is likely an easy fix if someone is looking at summary() anyway.

The following algebras would be nice to have. They are currently scheduled for completion after the 1.0 release:

ceiling, floor, trunk, round
dim
rowMeans, colMeans
cov2cor

The wiki page is out of date. I'll update it when I have a better interface than the iPhone. See data/omxSymbolTable.tab

The sorting of raw data is causing a bug when attempting to index using definition variables. Steve's script will move from my inbox to this bug report when I get back to civilization.

Steve has a test case where an integer type value causes the backend to barf. Calling as.numeric on the field should fix this problem. Writing this down so I don't forget.