Panel data and Integreated choice latent variable model (ICLV)

[kiran.t...@gmail.com]

Dear Prof. Bierlaire,

I am estimating the ICLV model on the panel data but encounter the following error during the iterations:

RuntimeWarning: overflow encountered in double_scalars rho = num / denom

What does this error mean?

And also in the results file, it says that the model terminated as the trust region was too small, that means it did not converge right?

Thanks for your help.

Kind Regards,

Kiran