refactor mcmc_intervals/mcmc_areas

[jgabry]

These are based on very old code and the same functionality could be provided using less code (and simpler code). Some improvements could be made too, for example using a discrete scale for the vertical axis.

[tjmahr]

Just some notes from studying the code.

Right now, we call mcmc_intervals() and mcmc_areas(), and that prepares a call to the workhorse function .mcmc_intervals(). That function prepares the data for both types of plots, creating the density curves based on a flag show_density if needed, and then produces the plots.

I will first factor out the data-preparation steps and export new functions mcmc_intervals_data() and mcmc_areas_data().

[jgabry]

Sounds great! These two functions are probably the worst in the package from a coding perspective, so there's a chance it will take less time to implement the functionality from scratch than to try to modify the existing code. If I were going to do it myself that's what I might do, but however you want to go about it is fine by me! On another note, since it would be good to eventually offer joy plots it might make sense to keep that in the back of your mind when working on the intervals and areas functions. I would think that a joy plot implementation could share some of the same internals.

…

On Fri, Sep 1, 2017 at 2:28 PM TJ Mahr ***@***.***> wrote: Just some notes from studying the code. Right now, we call mcmc_intervals() and mcmc_areas(), and that prepares a call to the workhorse function .mcmc_intervals(). That function prepares the data for both types of plots, creating the density curves based on a flag show_density if needed, and then produces the plots. I will first factor out the data-preparation steps and export new functions mcmc_intervals_data() and mcmc_areas_data(). — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#103 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AHb4Q2wmwvCQ2AuuurmFxRwzmDdGGoCHks5seEyygaJpZM4OwBQP> .