Cohort sequential design: preblem abo... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Jianhua Zhou posted on Monday, March 26, 2018 - 6:17 pm
Hi Linda,

I am also using cohort sequential design to obtain grade-based measurements to look at constructs across grade 3-5. Each individual was observed 4 times. I tried to do a LGCM and put some latent outcomes in tis model, but there always some problem about PSI. Could you give them some advice?

Here is the warning:

WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IN GROUP 1
IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/
RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL
TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE
THAN TWO LATENT VARIABLES. CHECK THE TECH4 OUTPUT FOR MORE INFORMATION.
PROBLEM INVOLVING VARIABLE PM.


PM is the latent outcome.
 Jianhua Zhou posted on Monday, March 26, 2018 - 6:22 pm
And here is my code:

DATA:
FILE IS 555.CSV;

VARIABLE:

NAMES ARE G r2-r19 n1-n4 g1-g4 z1-z4

jq ss jj sc se ho gu xq lo de sk ax;

usevari are G g1-g4 se jq de ax;

GROUPING = G (3=1 4=2 5=3);

Missing are all (99);

ANAL:STARTS=100;

MODEL:
pm by se jq;
nm by de ax;
pm @ 0;
nm @ 0;
i s |g1@0 g2@1 g3@2 g4@3;
[i] (1); [s] (2);
i (3); s (4);
i WITH s (5);
pm nm on i s;
g3-g4(6-7);
MODEL 2:
i s |g1@2 g2@3 g3@4 g4@5;
g1-g2(6-7);
g3-g4(8-9);

MODEL 3:
i s |g1@4 g2@5 g3@6 g4@7;
g1-g2(8-9);
g3-g4;
OUTPUT: TECH4 MODINDICES (ALL) STANDARDIZED;
 Bengt O. Muthen posted on Tuesday, March 27, 2018 - 11:58 am
This statement is strange

pm @ 0;
nm @ 0;

because it says there is no factor variance so the factor does not play a role.

Note also that we request that postings be limited to one window. For longer message, send to Support.
 Jianhua Zhou posted on Tuesday, March 27, 2018 - 7:24 pm
Thank you very much for your reply£¡

I have deleted the code:

pm @ 0;
nm @ 0;

but same warning is still there for all groups(1-3):

WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IN GROUP 3
IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/
RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL
TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE
THAN TWO LATENT VARIABLES. CHECK THE TECH4 OUTPUT FOR MORE INFORMATION.
PROBLEM INVOLVING VARIABLE PM.

I really don't know where the problem is.

Thanks for your help!
 Bengt O. Muthen posted on Wednesday, March 28, 2018 - 11:51 am
We need to see the full output - send to Support along with your license number.
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: