Total effects PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Angela Eakin posted on Wednesday, January 11, 2006 - 4:46 am
We want to be able to obtain the total effects. As an example we have used the following commands in the below model.

model indirect:
m4 ind i1 all3;
m4 ind s1 all3;

We have positioned this in different sections of the command lines but the below error message is frequently as follows:

*** ERROR
No valid keyword specified.


Title: all data 25.11.05;
data: file is all data prob differ stan and percent (2)..dat;

format is 10F8.0 3F8.4 1F8.0 7F8.4 56F8.2 1F3.2 14F8.2 8F5.2 9F8.4;
variable: names are class fsm schsize clsize ft3 con all3 only123 sex age
m1 pvn1 pvnvmc pvnvcat m2 m3 m4 r1 r2 r3 r4
pos1 hsv1 hio1 ind1 eff1 act1 epc1 tas1 per1
work1 self1 myte1 learn1 avoid1
pos2 hsv2 hio2 ind2 eff2 act2 epc2 tas2 per2
work2 self2 myte2 learn2 avoid2
pos3 hsv3 hio3 ind3 eff3 act3 epc3 tas3 per3
work3 self3 myte3 learn3 avoid3
pos4 hsv4 hio4 ind4 eff4 act4 epc4 tas4 per4
work4 self4 myte4 learn4 avoid4
books1 bdn1 bdn2 books3 bdn3 books4 bdn4
pv1 pv2 pv3 pv4 nverb1 nverb2 nverb3 nverb4
pvnv1 pvnv2 pvnv3 pvnv4 p1 p2 p3 cat
m1p r1p m2p r2p m3p r3p m4p r4p cm4p;
missing=blank;
cluster = class;

usevariables are m4 act1 act2 act3 act4 all3 only123 sex age fsm
class p1 p2 p3;
!missing=blank;
cluster = class;
classes=cg(3);
training=p1 p2 p3 (probabilities);
!knownclass=cg(cat=1 cat=2 cat=3);

analysis: type = mixture complex;
!starts=50 2;



model: %overall%
i1 s1 | act1@0 act2@1 act3*2 act4*2;
i1 s1 on all3 only123 sex age fsm;
i1 with s1;
i1 s1;

m4 on i1;
m4 on s1;

model indirect:
m4 ind i1 all3;
m4 ind s1 all3;

%cg#1%
i1* s1*;
i1 with s1;
i1 on all3 only123 sex age fsm;
s1 on all3 only123 sex age fsm;
i1 s1 | act1@0 act2@1 act3* act4*;
act1* act2* act3* act4*;
m4*;
m4 on i1;
m4 on s1 (1);
m4 on sex age fsm;



%cg#2%
i1* s1*;
i1 with s1;
i1 on all3 only123 sex age fsm;
s1 on all3 only123 sex age fsm;
i1 s1 | act1@0 act2@1 act3* act4*;
act1* act2* act3* act4*;
m4*;
m4 on i1;
m4 on s1 (1);
m4 on sex age fsm;



%cg#3%
i1* s1*;
i1 with s1*;
i1 on all3 only123 sex age fsm;
s1 on all3 only123 sex age fsm;
i1 s1 | act1@0 act2@1 act3* act4*;
act1* act2* act3* act4*;
m4*;
m4 on i1*;
m4 on s1;
m4 on sex age fsm;


plot:
type = plot1 plot2 plot3;
series = act1(1) act2(2) act3(3) act4(4);
output: sampstat standardized; modindices tech4 tech7 cinterval;
!savedata: file is 3class.sav;
!save is cprob;
 Linda K. Muthen posted on Wednesday, January 11, 2006 - 6:29 am
MODEL INDIRECT should not be part of the MODEL command. It is a separate command. Put it either before MODEL: or just before PLOT: If this does not work, send your input, data, output, and license number to support@statmodel.com.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: