Dear Professor Michel and everyone,
I am running a hybrid choice model with some latent variables.
When I estimated the model, some of the independent variables (latent variables) in the model have very huge coefficients (around hundreds) and were found to be significant. Could anyone explain why this is happening? and how to fix it?
Below is a part of the result from my model: