|
|
How to free the variance of model ? |
|
Message/Author |
|
Jyun-Ji Lin posted on Friday, April 20, 2007 - 12:36 am
|
|
|
I have a variable y, that y is the IRT result of 10101 students' ability value. Now I want to estimate the normal mixture model of y with 3 class latent model in the y. My M-PLUS program is followed: title: TASA 2005 standard setting cut score data: file is ab.dat; variable: names is y; classes =c (3); analysis: TYPE = MIXTURE; model: %overall% y*; output:tech1; THE MODEL RESULTS are followed: CLASS MEANS Y VARIANCES Y CLASS1 -0.863 0.347 CLASS2 -3.567 0.347 CLASS3 0.435 0.347 MY VARIANCE ALL ARE EQUAL 0.347. Can you help me to free the VARIANCES ? |
|
|
You free them by mentioning them in the class-specific parts of the model. |
|
Back to top |
|
|