repeated measures Lavaan regression model (with moderation)

[George]

I am attempting to set up a regression model with the following characteristics:

2 Categorical Independent Variables: X & Y

X has 3 levels

Y has 2 levels

[Subjects are tested at every level of both X & Y, hence repeated measures]

3 Dependent Variables O, P, T (all continuous)

1 Moderator Variable M (also continuous)

M moderates the relationship X & Y has on O, P, & T.

How do I set up a pure regression model with FIML in Lavaan that takes into account the repeated measures while also including the moderator in the syntax?

There are no latent factors.  All continuous variables are observed.

I would appreciate if anyone could point me in the right direction.  Thanks!

[George]

is this beyond the scope of Lavaan?  

[Terrence Jorgensen]

Since you have a continuous (rather than categorical) moderator, and you want a "pure regression" model, you can fit a multilevel regression

http://users.ugent.be/~yrosseel/lavaan/tubingen2017/multilevelSEM_part1.pdf

However, lavaan doesn't yet offer FIML for multilevel models.

Terrence D. Jorgensen

Postdoctoral Researcher, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

http://www.uva.nl/profile/t.d.jorgensen