The second derivatives is close to singularity

[Yuanjie Tu]

Hi everyone,

I am running an MNL, and the initial likelihood turns out to be the same as the final likelihood, and the cause of termination is "Relative gradient = 3.1e-17 <= 6.1e-06". Here is the result:

There are some error messages at the end of the report, saying "The second derivatives are close to singularity"

Do you know there is a way to solve this?

Many thanks for the help again!

Best,

Tukey Tu