I want to do bayesian regression in Python using PyMC3. The regression is in the following form: y = intercept + beta*function(data, parameters) + error Where the function embeds parameters (L, P, D) that are random variables I want to estimate and for which I initialize a prior distribution. Let…

Hey! It should be relatively easy to port your adstock function from numpy to aesara. See for example Media Effect Estimation with PyMC: Adstock, Saturation & Diminishing Returns - Dr. Juan Camilo Orduz :slightly_smiling_face:

Using custom python function within PyMC3

Questions v5

Matteo_Albrito August 2, 2022, 6:24am 3

Thank you! This is a great source for addressing my problem

Topic		Replies	Views
Geometric Adstock Paramter Estimation in PyMC3 and Theano (using theano scan) Questions theano	4	2301	September 9, 2018
Bayesian for simple ODE model. ValueError: Setting an array element with a sequence Questions	1	2389	July 28, 2018
Passing pymc3 distribution variable to theano function Questions	3	1649	September 23, 2019
Input parameter distribution of an open-source modelling code using pymc Questions	1	268	September 4, 2022
Re-coding a partial pooling from pymc to pymc3 Questions	1	467	November 2, 2018

Using custom python function within PyMC3

Related topics