Skip to content

Conversation

@dweindl
Copy link
Member

@dweindl dweindl commented Dec 2, 2025

Adds Poisson, binomial, and negative binomial distributions to the list of noise distributions.

Closes #653.

@dweindl dweindl force-pushed the discrete_noise_distr branch 2 times, most recently from a3fe413 to 8c9f5aa Compare December 2, 2025 11:47
@dweindl dweindl force-pushed the discrete_noise_distr branch from 8c9f5aa to ee15fcf Compare December 3, 2025 10:53
Adds Poisson, binomial, and negative binomial distributions to the list of noise distributions.
@dweindl dweindl force-pushed the discrete_noise_distr branch from ee15fcf to e94480b Compare December 3, 2025 10:56
@dweindl dweindl marked this pull request as ready for review December 3, 2025 10:59
@dweindl dweindl requested a review from a team as a code owner December 3, 2025 10:59
Copy link
Contributor

@sebapersson sebapersson left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for adding. Just a few minor comments.


The measured value in the same units/scale as the model output.
If the corresponding ``noiseDistribution`` specifies a discrete distribution,
this value must be integral.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
this value must be integral.
this value must be an integer.

the current specification assumes independence, i.e. the full distribution is

.. math::
\pi(D|Y,\Sigma) = \prod_i\pi(m_i|y_i,\sigma_i)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

As we have single-parameter distributions (e.g., Poison) that does not include a noise parameter, using a more general formula like: \pi(D|\theta) would maybe work better?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Poisson and negative binomial distributions for parameter priors and noise models

3 participants