Regression output: STDY vs STDYX PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Sakshi Bhargava posted on Monday, March 09, 2015 - 11:18 pm
Hi Dr. Muthen,
I am trying to run a regression model with both binary and continuous predictors in the same model. Can I read the STDY output to get the estimates of the binary predictors and the STDYX output for the estimates of continuous predictors OR is there another way to get the estimates of binary and continuous predictors?

Thank you,
Sakshi
 Linda K. Muthen posted on Tuesday, March 10, 2015 - 5:56 am
You should use StdY for binary predictors and StdYX for continuous predictors.
 Sakshi Bhargava posted on Tuesday, March 10, 2015 - 9:27 am
Thank you for responding so quickly. I just want to make sure that I understand what you mean.
1) Are you suggesting that in the final output that's generated, I read the StdY output for binary predictors and StdYX output for continuous predictors OR

2) Is there any way to specify in the syntax that I need Stdy for binary predictors and StdYX for continuous?

In my model I have
X1 = Gender (binary)
X2 = Race (binary)
X3 = SES (continuous)
X4 = Diversity (continuous)

My Model statement looks like:

Y1 ON X1 X2 X3 X4;
Y2 ON X1 X2 X3 X4;
Y3 ON X1 X2 X3 X4;
Y4 ON X1 X2 X3 X4;
Y5 ON X1 X2 X3 X4;

and in the output I mention:
OUTPUT: standardized;

Where in this syntax can I indicate that I need Stdy for binary predictors and StdYX for continuous.

I apologize in advance if I sound naive.

Thank you
 Bengt O. Muthen posted on Tuesday, March 10, 2015 - 12:27 pm
1)

2) No, you have to pick from different parts of the output.
 Bengt O. Muthen posted on Tuesday, March 10, 2015 - 12:28 pm
I meant to say

1) Yes.
 Sakshi Bhargava posted on Tuesday, March 10, 2015 - 12:38 pm
Okay. Thank you once again!
 Matthias Lühr posted on Monday, June 11, 2018 - 9:35 am
Dear Linda, dear Bengt,

we are conducting linear regression analyses (no mediators) with ML estimation. We are requesting stdyx-standardization as well as stdy-standardization in the output-command, but regression coefficients for stdyx and stdy are identical.

The formulas for stdyx and stdy provided in the User's Guide are:

StdYX = b*SD(x)/SD(y)
StdY = b/SD(y);

After using these formulas to calculate stdyx and stdy manually, we detected that the coefficients for both stdyx and stdy in the MPlus-output refer to the stdyx-formula. The stdy-coefficients that we calculated manually differ distinctively from the stdy-coefficients in the MPlus-output.

This problem is occuring in MPlus v8.0 and in MPlus v8.1.

Do you have any advice why stdyx- and stdy-coefficients are the same?

Thank you very much and best regards,

Matthias Lühr
 Bengt O. Muthen posted on Monday, June 11, 2018 - 5:46 pm
Perhaps your x's have been brought into the model by mentioning their variances. If this doesn't help, send your output to Support along with your license number.
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: