How to obtain saturated model fit?

3 replies [Last post]
wuhao_osu's picture
Offline
Joined: 09/07/2010

Hello, guys,

When using the MLObjective, how to obtain the saturated model fit? Is it somewhere in the output slot? If discrepancy function value can be found, that is even better. I think most softwares when analyzing a sample covariance matrix would return the discrepancy function value.

Thanks.

Hao

Ryne's picture
Offline
Joined: 07/31/2009
The saturated -2 log

The saturated -2 log likelihood is named "SaturatedLikelihood" in the output slot. Access it like so:

model@output$SaturatedLikelihood

wuhao_osu's picture
Offline
Joined: 09/07/2010
Thanks for the quick

Thanks for the quick response, Ryne.

I do not see it in the output slot, probably because I used a wrapper model to include two groups and the grand objective is the sum of the two group level ones. I checked the submodels but do not see it in fit@submodels$submodelA either.

Ryne's picture
Offline
Joined: 07/31/2009
Unless the groups are

Unless the groups are independent, the output should return to the top model rather than the group specific ones. As that objective is an algebra objective, the saturated likelihood isn't returned. We'll consider this a feature request.

It should be easy enough to put together a saturated model script for ML data. Let me know if you need a hand.