I’m trying to setup what in mplus would be called a parallel process model. It’s nothing new as I’ve seen it described somewhere, maybe by a different name, in the past 10-15 years, let’s say. The basic idea is to estimate two models within
single model statement. The data setup looks like this
Id DV ta tb inta intb
1 3 0 0 1 0
1 2 1 0 1 0
1 4 2 0 1 0
1 4 3 0 1 0
1 6 0 0 0 1
1 8 0 1 0 1
1 7 0 2 0 1
1 9 0 3 0 1
2 …..
Within each case the four records of series A and four records for series B are stacked in that order. I think the level one model is y = b0a*inta +b1a*ta + b0b*intb +b1a*tb + e.
The coefficient b0a is the intercept for series A and b0b is the intercept for series B. This is all an adaptation of how a growth curve is modeled and may not be right in a multilevel format.
The mixed statement for a random intercepts model would be
Mixed DV with inta ta intb tb/fixed inta ta intb tb/random inta intb | subject(id) covtype(un).
The problem is that I want to exclude the normal intercept term. There is keyword, NOINT, shown in the documentation but when I specify it in the fixed subcommand, I get an error. Usual questions: can mixed estimate such a model?, have
I made a mistake setting it up?, and if so, what is the correct statement?
Thanks Gene Maguin
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
