Intercept for latent variable PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Andreas Kissling posted on Thursday, March 22, 2012 - 5:56 am
Hi,

I have a model which structurally looks like this:

...
TYPE=RANDOM;
ALGORITHM=INTEGRATION;
MODEL:
f1 BY x1-x3;
f2 BY y1-y3;
f1sq | f XWITH f;
f2 ON f1 f1sq;

So this works fine and both f1 as well as the quadratic term f1sq load on f2. What I want to do now is draw a simple graph with a line given by a function like f(x)=a + bx + cx². I assume that the estimates that Mplus provides for regressing f1 and f1sq on f2 give me the b and c respectively. But I don't seem to be able to obtain an intercept, i.e. the "a" for my f(x). I succeeded producing a nice scatter plot from the data using the PLOT3 command but I would also like to show a simple curved line showing the relationship between f1 and f2.

Your help is greatly appreciated.
 Linda K. Muthen posted on Thursday, March 22, 2012 - 6:23 am
In a cross-sectional model, the intercept of a factor is zero.
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: