How to verify that uncertainty (estimated from pymc3) is accurate?

hd2005 · May 23, 2019, 4:37am

I have followed your above instruction and below is the results (M = 1028).

And here is the ECDF:

The above results are when I chose HalfGaussian Priors for my parameters. However, when Uniform Priors are chosen the U_i distribution looks “less” uniform. My naive conjecture is that HalfGaussian Priors are “better” suited for my problem than Normal Priors. Is that reasonable? FYI: below are results from Uniform Priors:

Topic		Replies	Views
Sampler occasionally gets static in pymc3.1, but not in pymc3.0 Questions development	19	2271	July 5, 2018
Uncertainty of Model Predictions Questions	4	2029	April 10, 2018
PyMC3-Sampling not taking place for the entire range of values between the given bounds of a Uniform prior Questions	7	1606	January 14, 2018
The 95% interval of PyMC3 trace doesn't cove the real values Questions	8	1863	July 21, 2017
Issue getting started with simple example Questions	1	729	November 18, 2018