SQM and Missing Values PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Boris Blechschmidt posted on Wednesday, December 19, 2007 - 7:53 am
Hello Drs. Muthen,

I try to compare different imputation techniques for missing values in three contexts: full SQM, CFA and PLS. Therefore I would like to estimate the 3 contexts x 2 sample sizes x 7 levels of missing values x 5 imputation techniques x 100 replications (21.000 analyses)
I did this before (just for a couple of estimations) by using STATA and AMOS.
With STATA it is possible to generate a dataset which belongs to a correlation matrix (the correlation matrix fits to a predefined SQM). I deleted several values, uses some imputation methods and estimated the paramteres with AMOS.
I’m not sure if Mplus (MONTECARLO) can make the same analyses with less effort.

Thanks,
Boris
 Linda K. Muthen posted on Wednesday, December 19, 2007 - 9:04 am
In Mplus Monte Carlo, missing data can be generated in two ways. They are described in Examples 11.1 and 11.2. You should be able to do what you want using the Mplus Monte Carlo simulation procedures.
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: