Error - NA values

1 reply [Last post]
Lucía's picture
Offline
Joined: 04/11/2013

Hi everybody,

I have been trying to run a model when checking the assumptions. It is a univariate analysis with an ordinal variable with two thresholds, and I have two covariates.

Everything seems to be fine, but I get this error message:

> baseFit <- mxRun( baseModel, intervals=FALSE )
Running base
Error: In model 'base' the thresholds in column 'BFEED12th_01' of matrix/algebra 'fMZ.expThreshfMZ' contain NA values. Only the first 2 elements of this column are inspected.

Do you know what may be happening?

Thank you

neale's picture
Offline
Joined: 07/31/2009
Guess

Without seeing the script I can only guess. It sounds like the threshold matrix has had values set to NA, or some matrix that multiplies it, or adds to it, contains NA's. Inspecting the objects in the model is a VERY useful diagnostic. Take a look at baseModel$fMZ.expThreshfMZ
for a start, then look at other relevant matrices that may contribute to it. You may even need mxEval() with particular data vectors if there are definition variables in some of the matrices.