Savedata for growth factors PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Sylvana Robbers posted on Friday, July 25, 2008 - 4:49 am
Hello,

I am wondering if it is possible to obtain growth factor estimates (intercepts and slopes from LGM) separately for each subject in the sample. I would like to save these values in a separate file. How can I do this? I already tried several things with the SAVEDATA-command, without success.

Thanks in advance.
Sylvana
 Linda K. Muthen posted on Friday, July 25, 2008 - 6:44 am
Use the SAVE=FSCORES option of the SAVEDATA command for this. If you continue to have problems, send your files and license number to support@statmodel.com.
 Sylvana Robbers posted on Friday, July 25, 2008 - 8:28 am
This worked, thanks alot! I am getting stars in the first column, and stars preceding some variables, but that isn't a problem. I am surprised that the growth factor scores are full numbers: I had expected scores like for instance -1,3549 instead of -1 for the slopes. Is this just my ignorance, or?
Thanks,
Sylvana
 Linda K. Muthen posted on Friday, July 25, 2008 - 8:51 am
The asterisks are missing values.

Please send your input, data, output, saved data file, and license number to support@statmodel.com.
 Sylvana Robbers posted on Friday, July 25, 2008 - 11:28 am
Weekend has started in Europe, so I'll send you the files on Monday. Thanks.
 Sylvana Robbers posted on Monday, July 28, 2008 - 3:00 am
Hello, I just solved the problem myself. Now I have three decimals for the intercepts and slopes of each subject.
Thanks!
 xiaoyu posted on Friday, May 02, 2014 - 4:21 pm
Dear Muthen,
For the conditional growth model, how to save the average estimated value for each time point, so that I can plot the graph in Excel?
Thank you,
 Bengt O. Muthen posted on Friday, May 02, 2014 - 5:11 pm
Just use the Save graph data option under the Plot menu.
 Brian Clark posted on Friday, February 27, 2015 - 6:03 pm
Is there any way to use individuals' correlations (via savedata maybe) between two growth/slope factors as a predictor, or more specifically in my case a mediator of group differences on an outcome? Is that legit?
 Bengt O. Muthen posted on Saturday, February 28, 2015 - 8:45 am
You can save all individuals' factor scores for the two slope factors, then correlate those factor scores, but that correlation won't vary across individuals.
 Brian Clark posted on Saturday, February 28, 2015 - 1:55 pm
Thanks! Sometimes you gotta ask a dumb question as a sanity check.
 Sabrina Krys posted on Wednesday, June 28, 2017 - 7:07 am
Hi,

I am trying to save intercept and slope parameters for a growth model using the Bayes estimator with FSCORES. However, I get the following error message:

*** ERROR in SAVEDATA command
The number of imputations must be specified when requesting FSCORES for ESTIMATOR=BAYES.

I am not sure what Mplus wants me to do, since I am not imputating any data?

Regards!
 Bengt O. Muthen posted on Wednesday, June 28, 2017 - 6:16 pm
See the V8 UG (on our website) page 838.

Imputation in this case refers to factor score estimation. That is, draws from a posterior distribution. The benefit of Bayes is that a distribution of values is obtained for each subject, not just the mean and SE.
 Sabrina Krys posted on Wednesday, June 28, 2017 - 11:41 pm
Thank you, that helped!
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: