Weights PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Anonymous posted on Wednesday, September 28, 2005 - 1:30 pm
I have a dataset that applies weights. My syntax is correct and my variables are read in correctly. However, I am getting a error that says, "Weight variable has nonpositive value at observation 1". I checked observation 1 and it has missing data for the first weight variable in obs. 1. Do all of the observations in the weight column have to have an observation?
 Linda K. Muthen posted on Wednesday, September 28, 2005 - 2:41 pm
Weight variables must be positive real numbers. Missing values are not allowed.
 Anonymous \ posted on Wednesday, September 28, 2005 - 3:01 pm
Thank you.
 jml posted on Wednesday, September 09, 2015 - 7:11 am
Hi,

I notice that in order to use weights along with the standard ML estimator, you have to enable bootstrapping. What is being bootstrapped here and why is it necessary to use bootstrapping?

Thanks,
Jon
 Linda K. Muthen posted on Wednesday, September 09, 2015 - 9:23 am
You do not need to enable bootstrapping to use weights with ML. What makes you think that?
 jml posted on Wednesday, September 09, 2015 - 10:20 am
Hi,

When I try running my syntax without a bootstrap option I get an error message that sampling weights are not available with ML and to use MLR instead. But it works fine when I add in bootstrapping.

Thanks,
Jon
 Linda K. Muthen posted on Wednesday, September 09, 2015 - 11:09 am
Please send the two relevant outputs and your license number to support@statmodel.com.
 susana barradas posted on Thursday, July 19, 2018 - 4:35 pm
I Linda,

I´m performing SEM analysys, and since I have sampling weights, I used WEIGHT IS in the my syntaxis. However, I keep receiving this error advice:

*** ERROR in ANALYSIS command
Unknown option:
WEIGHT

I´m now confused why is this happening, do you have any guess? Many thanks!!
 Linda K. Muthen posted on Thursday, July 19, 2018 - 6:04 pm
My best guess is that you don't have it in the VARIABLE command or you have intermingled two commands. If you can't see it, send the output and your license number to support@statmodel.com.
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: