| Message/Author |
|
|
|
|
Is it possible to extend the model below to incorporate TSCORES (to account for individual varying times of observation)?
If so does it matter if the age variables are centered or not?
ANALYSIS:
ESTIMATOR = MLR;
TYPE IS MIXTURE;
STARTS = 80 8;
MCONVERGENCE = 0.01;
ALGORITHM=INTEGRATION;
INTEGRATION = MONTECARLO;
PROCESS = 4;
MODEL:
%OVERALL%
i1 s1 | wt1@0 wt2 wt3 wt4 wt5 wt6@1;
i2 s2 | ht1@0 ht2 ht3 ht4 ht5 ht6@1;
i1 s1 ON sex bwt mage mwt;
i2 s2 ON sex bwt mage mht;
c ON sex bwt mage mwt mht;
bwt mage mwt mht; |
|
|
|
|
Q1: Should work - try it.
Q2: I assume you are not concerned about the mage variable but the time scores. The latter don't have to be centered. |
|
|
|
|
Thanks, I tried with the code below, but got multiple errors. Do you know where I am going wrong?
TSCORES = age1-age9;
ANALYSIS:
ESTIMATOR = MLR;
TYPE = MIXTURE RANDOM;
STARTS = 80 8;
MCONVERGENCE = 0.01;
ALGORITHM=INTEGRATION;
INTEGRATION = MONTECARLO;
PROCESS = 4;
MODEL:
%OVERALL%
i1 s1 | wt1@0 wt2 wt3 wt4 wt5 wt6@1 AT age1-age9;
i2 s2 | ht1@0 ht2 ht3 ht4 ht5 ht6@1 AT age1-age9;
*** WARNING in MODEL command
At least one variable is uncorrelated with all other variables within class.
Check that this is what is intended.
*** ERROR in MODEL command
Redefinition of the following random effects:
I2 S2 | w ` p AT AGE1 |
|
|
|
|
| Need to see your full output to conclude - send to Support along with your license number. |
|
|
| Back to top |
