|
Message/Author |
|
Steffi posted on Wednesday, July 26, 2006 - 3:47 am
|
|
|
Is it possible in Mplus to estimate a growth model that is not linear in the parameters, e.g. of the form Yit = 1+1/(1+ksi1*e^-(ksi2*Lambda) + eit (i.e., a logistic form when indicators are continuous)? THANK YOU! |
|
|
No. But see the literature by Brown and du Toit (Collins-Sayer longitudinal book on Best Methods) as well as Cudeck (MBR?). |
|
Back to top |
|
|