Growth curve modeling with multiple i... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Ana Velasquez posted on Thursday, November 05, 2009 - 11:15 am
Hi,

I am trying to model trajectories in aggression for a sample of over 1500 students. I assessed aggression 4 times over the course of a school year.

Given that I had missing data both in my outcome and covariates, I decided to do multiple imputation.

I followed the input model in you web training slides and example 12.13.

In running some preliminary analysis with Mplus I got two warning messages I am concerned about:

The first one is:

*** WARNING in PLOT command
No PLOT options are available for Monte Carlo analysis.
Request for all PLOT options will be ignored.

The second is:

THE CHI-SQUARE COULD NOT BE COMPUTED. THIS MAY BE DUE TO AN
INSUFFICIENT NUMBER OF IMPUTATIONS OR A LARGE AMOUNT OF MISSING DATA.


Any guidance as to how to get my data plotted and how to get fit indices would be greatly appreciated.

Thank you,

Ana
 Linda K. Muthen posted on Thursday, November 05, 2009 - 4:26 pm
If you need to use TYPE=IMPUTATION, you would need to plot outside of Mplus. The chi-square is computed in a special way for TYPE=IMPUTATION. It sounds like something about your data make this computation impossible.
 Daniel Hultell posted on Tuesday, June 25, 2013 - 4:44 am
Hello

I have a question about a LGM I'm trying to run with MI-data. The LGM is based on data from three waves and the variables are continuous. The model also includes a binary outcome that is regressed on the intercept and the slope. I also want to include four additional covariates to the model which causes some trouble since this reduces the sample size by approximately half. I have therefore imputed data for my four covariates (but not for the other variables in the model).

In the first step of the analysis I run the model without including the covariates, which should produce identical results regardless if the analysis is based on one dataset or on 100. When I specify the outcome as continuous the results are the same, but when I specify it as categorical the the effects of I and S differs completely (changes from positive and significant to a minimal negative effect and a p-value of .99). Since the outcome is binary I want present ORs for the effects of I and S. I have used MLR as estimator in both analyses

Do you have any explanation for why this happens and any suggestions how to resolve the problem.

Best regards

Daniel
 Linda K. Muthen posted on Tuesday, June 25, 2013 - 11:59 am
I don't understand what you are doing. Can you send the two outputs - continuous versus categorical and your license number to support@statmodel.com.
 Lina Homman posted on Friday, August 02, 2013 - 7:37 am
Dear Dr Muthen and Muthen,

I am having a problem with my plots when running a LCGA.
I am running a LCGA with 3 time points, outcome variables are continuous. I add covariates and direct effects from covariates to outcome variables. The model runs fine and the plots look good, but reduces my sample size. I decided to use MI for the covariates (I have tried to add them into the analysis but this does not work). The analysis runs fine but the plots do not. The issues with the plots is that the estimated means looks normal but all the sample means are consistently zero.
Do you have any idea where this is going wrong?
Many thanks

Syntax:
Usevariables= x1 x2 x3 y1-y9;
Classes = c (4);
Cluster=y9;
Data imputation:
Impute= y1 (c) y2 (c) y3 (c) y4 y5 y6 (c) y7 (c) y8;
ndatasets=50;
Analysis:
estimator=ml;
type=mixture complex;
Starts = 500 40;
Stiterations = 40;
Process=8;
Model:
%overall%
i s | x1@0 x2@1.5 x3*;
i-s@0;
c#1-c#3 on y1 y2 y3 y4 y5 y6 y7 y8;
x1 on y1 y3 y4 y5;
x2 on y1 y3 y5;
x3 on y1 y3 y6 y5;
Output:
SAMPSTAT STANDARDIZED tech1
tech8 tech11 tech14;
Plot:
Series=x1-x3 (s);
Type=plot3;
 Linda K. Muthen posted on Friday, August 02, 2013 - 8:34 am
If you are not using Version 7.11, I suggest that as a first step. If you have the same problem with Version 7.11, please send the relevant files and your license number to support@statmodel.com.
 Lina Homman posted on Friday, August 02, 2013 - 9:09 am
Many thanks for this Linda,

I have version 6. Which add ons would I need as to run the above model?

Many thanks
Lina
 Linda K. Muthen posted on Friday, August 02, 2013 - 9:42 am
You would need the Base plus Mixture or Base plus Combination Add-On.
 Lina Homman posted on Friday, August 02, 2013 - 9:45 am
great, many thanks:-)
 Lina Homman posted on Friday, August 02, 2013 - 9:53 am
Linda, I tested the Demo version 7.11 but it does still not work. The output looks fine it is just the plots which do not, the sample mean. The estimated mean is what I expect it to be. Do you have any other ideas on how to deal with this?
I cannot send the data files I am afraid due to data protection and confidentiality of the data.

Thanks for your help
Lina
 Linda K. Muthen posted on Friday, August 02, 2013 - 10:26 am
Please send the input and output files to support@statmodel.com. We will not be able to look into this until later in August when the programmer who handles this returns from vacation.
 Ashley Hum posted on Friday, June 06, 2014 - 5:04 pm
Hello,

Hello, I'm doing a GMM with individually varying times of observation and using 100 imputed datasets.

I would like to plot both the single-group and multi-class growth models, however, in an earlier post you said that for TYPE=IMPUTATION, you would need to plot outside of Mplus.

How could I get the information needed (e.g, means) to plot outside of Mplus?

Thank you,
Ashley
 Angela M. Stover posted on Saturday, October 18, 2014 - 5:42 pm
Hello,

Is there a way to request a Lo-Mendell-Rubin LRT in GMM with imputed data? I am getting an error stating that tech11 is being ignored because the data are imputed.

Thank you!
 Linda K. Muthen posted on Saturday, October 18, 2014 - 6:08 pm
This is not available for imputed data.
 DMello posted on Sunday, November 16, 2014 - 11:47 am
Hello!

After following the guidance of GCM using multiple imputation, my model ran successfully. However, the N was less than it should be if the data was imputed.

In other words, were the parameters estimated using the imputed datasets?

Here is my code:

TITLE: ...
DATA: ...
VARIABLE: NAMES = a1ct1-a1ct6 mincome mbod mlotr;
USEVARIABLES = a1ct1-a1ct6 mincome mbod mlotr
mincomeMbod mincomeMlotr mbodMlotr mincomeMbodMlotr;
MISSING = ALL (999);
DATA IMPUTATION:
IMPUTE = a1ct1-a1ct6 mincome mbod mlotr;
NDATASETS = 20;
SAVE = A1cImp*.dat;
ANALYSIS: ESTIMATOR = ML;
DEFINE:
mincomeMbod = mincome*mbod;
mincomeMlotr = mincome*mlotr;
mbodMlotr = mbod*mlotr;
mincomeMbodMlotr = mincome*mbod*mlotr;
MODEL: i s | a1ct1@0 a1ct2@1 a1ct3@2 a1ct4@3 a1ct5@4 a1ct6@5;
i s ON mincome mbod mlotr;
i s ON mincomeMbod;
i s ON mincomeMlotr;
i s ON mbodMlotr;
i s ON mincomeMbodMlotr;
OUTPUT: TECH1 TECH8;
 Bengt O. Muthen posted on Sunday, November 16, 2014 - 12:00 pm
Please send the full output and license number to Support, pointing to the differences in sample size that you refer to.
 Emma Davies posted on Wednesday, February 08, 2017 - 5:03 am
Hello,

I am conducting GMM with imputed data from STATA and I am preparing the files for MPlus. Will TYPE=IMPUTATION still work in MPlus if the imputations were done elsewhere? Also, the datafile with the 50 imputed datasets in it currently also contains the raw data (dataset 0). Does MPlus require the raw data to still remain in the imputed data file to give estimates, or would you advise this be dropped and only to use the imputed sets?

Thank you for your time.
 Bengt O. Muthen posted on Wednesday, February 08, 2017 - 4:10 pm
Look at Type=Imputation data is handled in UG ex 11.8 part 2 on our website:

http://www.statmodel.com/usersguide/chapter11.shtml

You see the imputed data sets and the needed "implist" file.
 Emma Davies posted on Sunday, April 02, 2017 - 3:26 pm
I notice in many posts that PLOTS are not available with TYPE=IMPUTATION and you have mentioned that you will need to plot outside of Mplus for this. I want to conduct my main analyses in mplus, but I guess I will need to find an alternative for graphical representation, as I am using imputed data. Do you have any recommendations of what other package might be best for this? Do you know of any tutorials you can point me towards?

I want to plot the estimated v sample means and am currently trying to do this in R, but I'm not sure it can be done and would really appreciate some guidance!

Thank you.
 Bengt O. Muthen posted on Thursday, April 06, 2017 - 5:48 pm
I would keeping trying using R. We have some R plotting setups on our website. See

http://www.statmodel.com/mplus-R/
 Cherry Chu posted on Wednesday, November 08, 2017 - 6:38 pm
Hi,

I want to create plots of estimated means using TYPE=IMPUTATION, and I see in the post above that R is recommended to plot outside of Mplus. However, the webpage says that a GH5 file will only be created with the PLOT command, but the PLOT command is ignored when TYPE=IMPUTATION is used. How else can I get plots of imputed data?

Thank you.
 Linda K. Muthen posted on Thursday, November 09, 2017 - 6:10 am
You will need to put together the information you want to plot and use R or another plot program.
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: