Changing a parameter that isn't on the chain every sample

ARB · April 1, 2021, 12:24pm

Sorry for the late reply. Wouldn’t using A_noisy = pm.Normal("A_noisy", A, 1) allow the sampling to change the distribution of A over the chain? I would want my posterior distribution for A to be the same as its prior.

Topic		Replies	Views
Building a normal likelihood with known but varying variance for every data point/sample Questions	1	420	February 20, 2021
Marignalization of Nuisance parameters Questions	3	963	January 19, 2018
Pymc3 evaluates function but doesn't update sample Questions	5	1369	November 3, 2017
Adding node to pymc3 model breaks sampling Questions	1	610	February 1, 2018
How to change Posterior distribution after sampling some draws Questions	1	478	November 22, 2018

Changing a parameter that isn't on the chain every sample

Related topics