pymc-devs/pymc

Implement DiffTransform for RandomWalk distributions

Open

#6.098 aberto em 4 de set. de 2022

Ver no GitHub
 (1 comment) (0 reactions) (0 assignees)Python (1.902 forks)batch import
enhancementshelp wantedpytensor

Métricas do repositório

Stars
 (7.926 stars)
Métricas de merge de PR
 (Mesclagem média 11d) (7 fundiu PRs em 30d)

Description

My guess is that unobserved [Gaussian/MvNormal/MvStudentT]RandomWalk would sample much better if NUTS could propose innovation values instead of the "absolute" timeseries values. This is what is done when one manually creates a random walk model by

x_raw = pm.Normal("x_raw")
x = pm.Deterministic("x", x_raw.cumsum())

If we just added the transform, there would some redundancies in the logp, because the proposed diffed values would be cumsumed (in the back-transform) and then differentiated again in the randomwalk logprob, but we can probably teach Aesara to optimize those away.

Once we have that, the user won't be worse off when creating unobserved random walks via the specialized PyMC distribution classes.

Guia do colaborador