importance sampling and hyperparameter methods pymc

时间:2018-02-03 09:50:17

标签: python statistics bayesian pymc

I am learning how to programing in Python and I am new using PyMC. I am trying to programing my own code for a bayesian parameter inference. My problem is that I need to introduce 2 "sections" to my code but I am not sure if PyMC can do it. For example, the topics that I need are:

Introduce an hyperparameter method, which is a method used when different datasets need to be taken into account. The method can be found here: https://arxiv.org/pdf/astro-ph/0203259.pdf. Is there something similar implemented on PyMC? Or what tools could I use if I would like to programing it?

The second "section" is about importance sampling. As you can see in this article, "https://arxiv.org/pdf/astro-ph/0205436.pdf": "Assuming that one has some new data which is broadly consistent with the current data, in the sense that the posterior only shrinks, one can use importance sampling to quickly compute a new posterior including the new data..." . Can I do this in PyMC?

Thanks for your time.

LP

0 个答案:

没有答案
相关问题