MPML estimation method PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
Message/Author
 Caroline Vandenplas posted on Thursday, April 08, 2010 - 1:58 am
Hello,
We would like to know whether the MPML estimation method is automatically applied in Mplus if we run a multilevel model using Monte Carlo simulated data and specify a cluster sampling design with weights. We cannot find a possibility to predefine this estimation method in the user guide.
The method is described in Asparouhov, T. & Muthén, B., 2006. General multi-level modeling with sampling weights. Communications in Statistics: Theory and Methods, 35(3), 439-460.)
Kind regards,
Caroline & Sabine
 Linda K. Muthen posted on Thursday, April 08, 2010 - 8:04 am
Mplus can generate data with clustering but not with weights. Mplus can analyze multilevel data with clustering and weights.
 Caroline Vandenplas posted on Tuesday, May 11, 2010 - 4:45 am
Dear Linda,

We figured out a way how to generate and attach weights to the datasets originated from Monte Carlo. Then, we analyzed these datasets in various multilevel models (specifying clusters and weights). The output gives "Estimator = MLR". We would like to know in what circumstances the MPML estimation method would be applied or how we can specify that Mplus would use this method for estimation.

Thanks a lot,
Sabine & Caroline
 Linda K. Muthen posted on Tuesday, May 11, 2010 - 9:44 am
MLR does MPML. You don't need to anything more.
 Jak posted on Monday, May 16, 2011 - 8:34 am
Dear Linda or Bengt,

Is MLR also equivalent with MPML if you do not supply weights?

I am trying to figure out what MLR estimation implies.

Thanks in advance!
 Tihomir Asparouhov posted on Wednesday, May 18, 2011 - 10:11 am
Yes, MLR is the same as MPML without the weights.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: