How can one fixed $\zeta$?

[digdig]

Is there any place to set the bounds of $\zeta$ to a fixed value? For example, I want set $\zeta=1$ as in many cases. In particular, in the autoscale method

[JefferyWangSH]

I have added three optional arguments fix_rho_c, fix_nu, and fix_zeta to the autoscale method. With these options setting to True, the optimization will proceed with corresponding parameters fixed as their initial guesses. The updated source code can be found here.

Also note that a bad guess shall lead to poor quality of optimization especially when parameters are fixed, e.g. sometimes the associated errors can be evaluated to be nan. So be cautious when use.