Single Stochastic Imputation for Miss... PreviousNext
Mplus Discussion > Missing Data Modeling >
Message/Author
 Carlo Pulcini posted on Monday, June 20, 2016 - 12:53 am
Dear Professors,
I have a question regarding Stochastic Imputation of missing data.
I want to perform a single (not multiple) imputation of missing data (coded -99 in my dataset).
I took inspiration from chapter 11 and in particular from the example 11.5 to write the following code

--TITLE: Stochastic simple imputation;
--DATA: FILE IS ***;
--VARIABLE: NAMES = Var1-Var48;
--
-- USEVARIABLES = Var2-Var48;
--
-- AUXILIARY = Var1;
-- MISSING = ALL (-99);
--
--DATA IMPUTATION: IMPUTE = Var2-Var45;
-- NDATASETS = 1;
-- SAVE = ***;
--ANALYSIS: TYPE = BASIC;

My question is: is this a correct way to perform single imputation (using the procedure for multiple imputation but with NDATASETS = 1)?
Is there a better way to perform single stochastic imputation in mplus?
Same questions if some of the variables to impute are categorical (of course adding the line CATEGORICAL in VARIABLE command and the (c) in IMPUTE)
Thank you in advance.
Carlo Pulcini
 Bengt O. Muthen posted on Monday, June 20, 2016 - 4:35 pm
Q1. Yes.

Q2. No.

Q3 - . Same answers.
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: