

Bidirectional model with 1 occasion m... 

Message/Author 


I want to make a bidirectional model of 3 variables (A, B and C) of which 1 is only measured twice: lets assume A is modeled twice at T1 and T3 and B and C are measured all 3 times. Is it possible to model this, or will this model be unstable? I'm especially worried about the following crosslagged pahts: A1 > C3 OR B1 > A3 A1 > A3 Or is it better to only take the paths at the following occasion (from 1 to 2 and avoid paths from 1 to 3) 


If A is not measured at time point 2, there is no variable A at that time point. The data set would be A1 A3 B1 B2 B3 C1 C2 C3 You can model the above variables. It looks like you don't have A2 in the model above. 

Back to top 

