I am trying to expand the model in the page https://docs.pymc.io/en/v3/pymc-examples/examples/generalized_linear_models/GLM-binomial-regression.html by adding a log normal noise to the Binomial distribution. I feel like this might be problematic as I am adding a continous RV to a discrete one. Nev…

[image] iavicenna: llike = pm.Lognormal("llike", mu=mu + pm.math.log(n*p), sigma=sd, observed=data) No, I meant the original form. The mean of the Lognormal is the mean of a latent Normal that is then exponentiated. Conversely, the log of a Binomial expectation would be log(n * p) or log(n) +…

Binomial Regression with Lognormal Noise

Questions v3

iavicenna January 26, 2023, 3:52pm 7

Thanks for the insight and the answer, it was quite helpful.

2 Likes

Topic		Replies	Views
LogitNormal vs. Beta vs. Logistic Questions	1	1089	August 15, 2018
Poisson realization of log-normal distribution Questions	4	873	January 28, 2021
Adding a known amount of random error to a model Questions	2	487	May 19, 2019
How to model observed percentages (bounded from 0 to 1) Questions	8	2801	January 3, 2018
Samples from prior appear to have wrong distribution Questions	2	606	October 8, 2018

Binomial Regression with Lognormal Noise

Related topics