Variables selection methods , minimum number of variables - max information

[Chiara Tasselli]

Good morning to everyone,
I am trying to run a SEM with more levels: the fact is that it is very heavy and I would like to reduce variables in the best possible way.
For now, as I have a lot of variables which explain one dimension, I simply try to look at the significance of coefficients and the increase of R² (in normal regression) contributed by each individual regressor.
I was wondering if there is a more precise/sophisticated methodology (inside or outside SEM) ?

Many thanks in advance for your time,
wishing you all a great weekend ahead!

[Andrew Musau]

This is a broad topic and suggestions may include "use those variables that are commonly used in the literature" to machine learning algorithms such as principal component analysis and lasso. For model selection using lasso, see

help Lasso_intro