|
|
Constrain variance and covariance to 0 |
|
Message/Author |
|
Anh Hua posted on Friday, June 19, 2020 - 6:58 am
|
|
|
Hello! How do I constrain the variance of S3 and the associated covariances (s3 with s1; s3 with s2) to 0? I've browsed through the mplus user guide and the forums and tried by adding s3@0 at the end of my MODEL syntax but this did not work. Thank you. Here is my syntax below: title: your title goes here data: file = "q2.reading.comp.mplus.final.dat"; variable: names = child months months_c test_rit_score; usev test_rit_score months_c months2_c months3_c; missing=.; within = months_c months2_c months3_c; cluster = child; Define: test_rit_score=test_rit_score; months2_c=(months_c*months_c)/5; months3_c=(months2_c*months_c)/5; analysis: type = twolevel random; estimator = mlf; iteration = 20000; miterations = 20000; convergence = .000001; algorithm = em; !this was added based on mplus error and suggestion. model: %within% s1 | test_rit_score on months_c; s2 | test_rit_score on months2_c; s3 | test_rit_score on months3_c; %between% s1 with test_rit_score; s2 with test_rit_score; s3 with test_rit_score; s1 with s2; s1 with s3; s2 with s3; |
|
|
Don't make the s3 slope random; just say test_rit_score on months3_c; |
|
Anh Hua posted on Sunday, June 21, 2020 - 7:59 am
|
|
|
Dr. Muthen, Thank you so much for your helpful advice! I'm a complete mplus newbie and your advice really helps me get better at modeling in mplus! |
|
Back to top |
|
|