You are here

Missing CIs - How to get them?

6 posts / 0 new
Last post
YiTan's picture
Offline
Joined: 10/08/2014 - 05:28
Missing CIs - How to get them?

Hi all,

When I ran the univariate or bivariate scripts for some of my variables, I often get NA for some of the CIs requested, followed by this line: "Investigate missing CIs with the detail output in MxComputeConfidenceInterval "

I would like to know what causes the missing CIs and what can I do to get the missing CIs? I'm not sure if I understand what arguments to put in the MxComputeConfidenceInterval.

Thank you very much for the help!

Best Regards,
Yi Ting

jpritikin's picture
Offline
Joined: 05/24/2012 - 00:35
CIs

After mxRun, you can find the diagnostics in model$compute$steps$CI$output$detail

Typically, the reason you get NAs is because your model isn't sufficiently identified.

YiTan's picture
Offline
Joined: 10/08/2014 - 05:28
NULL?

Hi Joshua,

Thanks for your reply, unfortunately, I got NULL when I input 'AceFit$compute$steps$CI$output$detail'. Did I miss something?

AdminRobK's picture
Offline
Joined: 01/24/2014 - 12:15
summary with verbose=T

If you're running version 2.2.x, I think you should be able to see the detailed CI output if you do something like summary(myFittedModel,verbose=T).

jpritikin's picture
Offline
Joined: 05/24/2012 - 00:35
version?

What does mxVersion() say?

YiTan's picture
Offline
Joined: 10/08/2014 - 05:28
mxVersion

Hi Joshua,

Here are my mxVersion details:

OpenMx version: 2.2.4 [GIT v2.2.4]
R version: R version 3.2.1 (2015-06-18)
Platform: x86_64-w64-mingw32
Default optimiser: SLSQP