Latent Variable Estimation

[Daniel Davis]

For the mirt function if you put a regression equation into the formula part what is it doing differently than estimating latent variable scores and then running a regression separately? To put this question another way I know you can theoretically run a regression with the whole distribution of a latent variable instead of just point estimates, like a MIMIC model, is that what this does? Thanks.

[Phil Chalmers]

Yes, it's exactly like what the MIMIC model does. The estimation of the parameters is baked-in to the likelihood equations, which is why likelihood ratio tests are also possible in mirt. 

Phil

On Sat, Dec 2, 2017 at 6:06 PM, Daniel Davis <iamdanth...@gmail.com> wrote:

For the mirt function if you put a regression equation into the formula part what is it doing differently than estimating latent variable scores and then running a regression separately? To put this question another way I know you can theoretically run a regression with the whole distribution of a latent variable instead of just point estimates, like a MIMIC model, is that what this does? Thanks.

--
You received this message because you are subscribed to the Google Groups "mirt-package" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mirt-package+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Daniel Davis]

Thanks that helps a lot.

Phil

To unsubscribe from this group and stop receiving emails from it, send an email to mirt-package...@googlegroups.com.