stan-dev / stan-dev/bayesplot

mcmc_pairs bivariate off-diagonal plot axis incorrect.

Open
#132 6 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug help wanted
Dominant language
R
Stars
442
Forks
93
Avg merge
3h 11m
Merged PRs (30d)
1

Description

Data file: https://drive.google.com/open?id=1pnAUDCz9uscccSaUISodF9cqGF23uNzU

Off-diagonal bivariate scatter plot axes are out of line

My understanding from the documentation is that half the chains are used for each of the off-diagonal plots. In this example case (see posterior data file linked above) the chains have not converged and so the lower off-diagonal plot seems to be using chains that only sampled from one mode, using default condition I think this should be correct. I believe the x-axis should be inline with the above univariate marginal plot and the y-axis inline scaled as the x-axis of the upper off-diagonal bivariate plot?

load(file = "posterior.arr.RData")
mcmc_pairs(posterior, pars = c("u","v"))

mcmc_pairs_joint_samples_missing_data

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Reproduce the report with posterior.arr.RData and the shown mcmc_pairs(posterior, pars = c("u","v")) call. Start at the mcmc_pairs plotting entry point and compare the lower off-diagonal axes with the upper bivariate plot and univariate margins. Done means the axis ranges and alignment match the reported expectations for the non-converged chains.

Written by the indexing model from the issue text.

Assessment

Tech stack
r
Domain
data-visualization
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.