Autoregressive model vs autocorrelate... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Johnny Wu posted on Tuesday, January 30, 2007 - 5:41 pm
Hello Dr. Muthen,

Say we had y1-y4:

Autoregressive models would regress y2 on y1, y3 on y2, and y4 on y3.

Autocorrelated disturbances would have the residual variances of y1-y4 be related to each other (given a latent growth model framework, where an intercept and slope is estimated from y1-y4 scores).

Substantively, how are they different?

 Bengt O. Muthen posted on Tuesday, January 30, 2007 - 6:01 pm
The former model regresses all of y on earlier y's, the latter model regresses only the residual part of y on earlier residual parts.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message