SEM related_" ******* "sign... PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Sanjoy posted on Friday, June 03, 2005 - 2:33 am
Dear Professor/s ... this is my model
R by R1-R3;
B by B1-B3;
Y on R B X1;
R on B X2;
B on R X3; (X's share some common elements, Ri's and Bi's are 5 point ordered categorical and Y is 0/1)

I ran the model ...there is NO error message of any type, I checked twice, still I'm getting " ***** " under SE column, ... I asked for "TECH 3", from the ESTIMATED COVARIANCE MATRIX FOR PARAMETER ESTIMATES, I found for some parameters the value is exceptionally high and for some of them it's " *****"

Could you please tell me some possible remedies and why is it happening? ...thanks and regards
 Linda K. Muthen posted on Friday, June 03, 2005 - 12:47 pm
This is a support question. Please send output, data, and your license number to support@statmodel.com.
 Sanjoy posted on Friday, June 03, 2005 - 6:46 pm
Ok Madam ...I am going to send, thanks
 Bettina Doering posted on Wednesday, August 05, 2009 - 10:47 am
Dear all,

I have the following model with manifest categorical variables.

Model:
w2_taet1 on w1_opf1 w1_taet1;
w2_opf1 on w1_opf1 w1_taet1;
w3_taet1 on w2_opf1w2_taet1;
w3_opf1 on w2_opf1 w2_taet1;

w1_opf1 with w1_taet1;
w2_opf1 with w2_taet1;
w3_opf1 with w3_taet1;

If I have not the statement "categorical are" the estimation works well. With the statement, SE are not estimated for some variables. My question is, what is the fault in the categorical model and what statement is missed?

Thank you for your help,

Bettina
 Linda K. Muthen posted on Wednesday, August 05, 2009 - 12:06 pm
Please send your input, data, output, and license number to support@statmodel.com.
 Bettina Doering posted on Friday, August 07, 2009 - 10:48 am
I have forgotten to write that the categorial variables are dichotomous. Could it be that the problem arises when I have a latent dependent variable which is predicted by dichotomous independent. The SE are not estimated than.

Thank you for your first answer and I am looking forward to hear from you.

Bettina
 Linda K. Muthen posted on Friday, August 07, 2009 - 2:05 pm
I don't think this would be the problem. Please send your input, data, output, and license number to support@statmodel.com.
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: