Imputation and memory limits PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Patrick Malone posted on Saturday, December 22, 2007 - 2:04 pm
Hello.

I'm still waiting on my 64-bit OS upgrade, so meanwhile I'm fighting a lot of memory battles.

I'm running a difficult model with numeric integration (5^4 points) on imputed data.

The first dataset seems to run normally, but the second one runs out of memory. Is this likely to be a quirk of the second dataset, or does the memory requirement of the problem cumulate across datasets?

Thanks.
 Linda K. Muthen posted on Saturday, December 22, 2007 - 4:55 pm
Please send the input, data sets, output, and your license number to support@statmodel.com.
 Kristine Khachatryan posted on Thursday, November 24, 2011 - 4:49 pm
I am trying to impute my data for multiple group analysis (n=2016). There are missing values on x and y variables. The y variables are categorical, the x variables are continuous
or dummy-coded. In the syntax i do not declare the y variables as categorical, but use the option „values“ to assign their values. The imputation runs fine. But when I try to impute using the grouping option in addition, i get the error message „THERE IS NOT ENOUGH MEMORY SPACE TO RUN Mplus ON THE CURRENT INPUT FILE...“. Is there anything wrong with my syntax or it’s only a memory space problem? What would you suggest to solve this problem? Would it be a too rough approximation to leave out the option "grouping" when doing the imputation and instead use it on the second step with the already imputed data? I very much appreciate your help.

Variable:
NAMES ARE ....;
MISSING ARE all (-99);
GROUPING IS REGION (1=OST 2=WEST);

DATA IMPUTATION:
Impute = v_lokpol-v_gricht eduyear l_einkom wrtlage perf_fk-demidee rechts links
egwrtlag reldprv rgrparty-erwrbls autonomy;

Values=v_lokpol-v_gricht (1-5) wrtlage (1-5) perf_fk-demidee (1-4)
rechts-links (0-1)egwrtlag (1-5) reldprv (1-4) rgrparty-erwrbls (0-1);

NDATASETS = 10;
SAVE = Agaube_imp*.dat;

ANALYSIS: TYPE = BASIC;
OUTPUT: TECH8;
 Linda K. Muthen posted on Saturday, November 26, 2011 - 8:23 am
The way you have your input set up, all variables on the NAMES list will be used to impute the data. Please see the latest Topic 9 course handout where there is a newer example of how to use NAMES and USEVARIABLES options with imputation. If this does not help, send the relevant files and your license number to support@statmodel.com.
 Kristine Khachatryan posted on Monday, November 28, 2011 - 1:37 am
It was very helpful, thank you very much!
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: