|
|
Categorical outcome and time-invaryin... |
|
Message/Author |
|
Yang Xiao posted on Monday, November 29, 2010 - 3:19 pm
|
|
|
Dear Drs. Muthen, I am trying to fit the trajactory of the categorical outcome variable defined as smoking status(csmk02-csmk04),which are ordered from 1-4,measured at year1 up to year3. Below is my program: DATA: FILE = 'Z:\NSPY\psu_m1.txt'; VARIABLE: NAMES = yraceeth csmk02 csmk03 csmk04 ; usevariables = csmk02 csmk03 csmk04; class = c(3); !CATEGORICAL = yraceeth; Model: %Overall% i s |csmk02@0 csmk03@1 csmk04@2; !i s ON yraceeth; Analysis: Type=mixture; Starts = 100 10; Output: tech1 tech11 ; Plot: type = plot3; series = csmk02 csmk03 csmk04(*); Now I want to see whether the trajactory is different among the three racial group(yraceeth),which is a time-invarying covariate.Could you provide an similar example in the manual of version 5? BTW,I tried to run the code including the comments as above,which does not work. Thanks! Yang |
|
|
If you want to use the three-category variable yraceeth as a covariate, you need to create two dummy variables. You can then regress c on the two dummy variables to see if class membership is related to race and regress i and s on the two dummy variables to see if the trajectory is related to race. If you have a problem with an input file, please send the problem and your license number to support@statmodel.com. |
|
Back to top |
|
|