"type=missing" and MLM PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Yan Li posted on Friday, December 08, 2006 - 9:47 am
I was trying to use both "estimator=MLM" and "type=missing H1" at the same time and I got this error message:

" Estimator MLM is not allowed with TYPE = GENERAL MISSING.Default will be used."

Can you let me know what's going on? Thank you!

Below is my syntax:

Usevar are
tr1-tr7 tr16 tr17 pn9 pn10 pn12
tr8-tr11 pn6-pn8;


analysis: type=missing H1;

SA by tr1-tr7 tr16 tr17 pn9 pn10 pn12;
OA by tr8-tr11 pn6-pn8;

SA with OA;

output: stand;sampstat; mod (20);
 Linda K. Muthen posted on Friday, December 08, 2006 - 10:32 am
We no longer allow MLM and MISSING. We recommend MLR and MISSING instead if you want a robust estimator.
 Yan Li posted on Monday, December 11, 2006 - 8:38 pm
Thank you Dr. Muthen. Here is a follow up question. I want to do a model comparison for nested and comparison models. I read your guidance about "Difference Testing Using the Loglikelihood" on the website. One thing confuses me is that I got L0 and L1 for both models (2 L0 and 2 L1 values) and I don't know which L0/L1 to plug into the formula (TRd = -2*(L0 - L1)/cd).

(2) Also does the parameters p0/p1 refer to the # of free parameters in the output?

(3) At last, when we have the TRd, shall I look at the chi-square table using df difference (nested df-comparison df) and check out the p value?

I appreciate any help you can give. Thank you!
 Linda K. Muthen posted on Tuesday, December 12, 2006 - 3:01 pm
1. L0 is the more restrictive model. L1 is the less restrictive model. You want the H0 loglikelihood from L0 and L1. The H1 loglikelihoods are not needed.

2. Yes.

3. Yes.
 RuoShui posted on Sunday, September 29, 2013 - 10:25 pm
What estimator should I use if I have data that has high skewness and kurtosis? I heard that MLM is the one to choose. Is this correct? Thank you!
 Linda K. Muthen posted on Monday, September 30, 2013 - 6:36 am
Both MLM and MLR are robust to non-normality. With MLM, listwise deletion of any observation with a missing value on one or more analysis variables is used. MLR uses all available information.
 RuoShui posted on Monday, September 30, 2013 - 7:02 am
I see. Thank you so much. I wonder, if I have categorical indicators as well as continuous indicators with skewness and kurtosis issues, should I still use MLR? Or do I need to use WLSMV?

Thank you.
 Linda K. Muthen posted on Monday, September 30, 2013 - 10:44 am
WLSMV is not robust to non-normality for continuous variables. You can use MLR with a combination of continuous and categorical indicators. Factors with categorical indicators require numerical integration so we recommend not too many of these factors.
 Yessenia Castro posted on Monday, November 25, 2013 - 9:41 am
Under similar circumstances listed by RuoShui above (a model with continuous predictor variables, one of which exhibits kurtosis, and a categorical outcome variable), is MLR still appropriate if the categorical variable is in fact binary?
 Linda K. Muthen posted on Monday, November 25, 2013 - 12:31 pm
In regression, the model is estimated conditioned on the observed exogenous variables. No distributional assumptions are made about them. You can use WLSMV or MLR with a binary dependent variable. Distributional assumptions are made about only continuous dependent variables.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message