Regression with multiple dependents PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Sandra Lyons posted on Wednesday, February 10, 2010 - 2:09 pm
I want regress a 13dependent variables (5 continuous and 8 dichotomous)on a set of 24 dependents with missing data. What is the advantage of using Mplus simultaneous regression over OLS/Logit in some other software? For example: (a) adjusted p for multiple tests, (b) smaller standard errors because correlated error terms are taken into account, (c)other?
 Bengt O. Muthen posted on Wednesday, February 10, 2010 - 4:58 pm
Those are relevant - and with missing data on the DVs, for any given regression the simultaneous analysis helps by drawing on information from the other DVs not missing for that subject (under MAR).
 Sandra Lyons posted on Wednesday, February 10, 2010 - 9:21 pm
Thank you. I'm looking forward to the Mplus short course in March.
 Paul A.Tiffin posted on Thursday, February 25, 2010 - 1:17 pm
Dear mplus team
I have a quick query
How is missing data handled in a straight forward regression, such as in path analysis with continuous variables, if the LISTWISE option in DATA is not used?
Many thanks for your help
 Linda K. Muthen posted on Thursday, February 25, 2010 - 2:02 pm
Mplus provides maximum likelihood estimation under MCAR (missing completely at random) and MAR (missing at random; Little & Rubin, 2002) for continuous, censored, binary, ordered categorical (ordinal), unordered categorical (nominal), counts, or combinations of these variable types.
 Paul A.Tiffin posted on Thursday, February 25, 2010 - 5:46 pm
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message