Degradation platformHi,
A customer is asking the following:
First of all see please below data obtained by using “Degradation data Analysis” in JMP. Have then two questions:
1. On the non linear regression path, which iterative algorithm is used
2. To get the convergence on the information criteria (AIC and BIC) which parameter is used and at which threshold?

Can you help?
Thanks,
Valerie
Re: Degradation platformHi Valerie,
The Standard now is BFGS, a Newton Based Optimization algorithm which is very capable and stable from my experience. I used it in my diploma theses as well.
And here is the answer from TS:
+++++++++++++++
Here is the response I got from the developer. They were purposely removed. Based on user experience that I have got since release of the platform. User usually has difficulty to select and configure optimization methods. For now, the underlying optimizations all use BFGS, a general purpose optimization routine, which should be sufficient to handle most if not all situations. Existing scripts should run. If not, please let us know. Please let me know if you have follow-up questions, or need clarification.
+++++++++++++++
Best, Martin
Re: Degradation platformLaura Archer from US TS a opened a bug Report with the development team. Stay tuned
Re: Degradation platformHi Valerie,
it seems that in JMP 10 and 11 the method and optimization button are there, but in 12 it isn't. So there must have been a reason for this Change. I'm still waiting for the answer from TS.
Martin
Re: Degradation platformHi Valerie,
in the documentation you will see several algorithms mentioned like Newton, QuasiNewton SR1 and 2. They also mention there is an optimization button where you can select this, but I couldn't find it in the platform. I have send this to Support to get an answer to that optimization method and why the button is not there but mentioned in the documentation. I guess it was in the tool in earlier Versions.
The question now is what is the Default used since one cannot Change it.
If I got new Information I'll let you know.If anyone else knows the answer to that please provide your Input