Mette Ranta posted on Wednesday, December 02, 2009 - 4:49 am
Hi, I am using a 4-measurement point latent growth model of longitudinal data. I receive the following problem: WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) 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 I. Could the problem with variable 1 be that it has over 100 missing variables, which I have marked as 999? The other variables have missing info too, but not to this large extent. I am happy to send you more information, thank you for your help!
This usually means that the variance of i is negative or that i is correlated greater than one with another variable. You can see this in the TECH4 output. If this does not resolve your problem, please send the output and your license number to firstname.lastname@example.org.