Consider integrating with stantargets

[martinmodrak]

The stantargets package seems to have some implementation of SBC: ropensci/stantargets#31

it might make sense to either allow using stantargets to manage fits and/or allows SBC package to be used from within stantargets. Not sure if either is sensible, just noting the possibility.

[hyunjimoon]

How big is stantarget package?
By the way what do you think about "not sure I would trust variational Bayes for SBC" from the link?

I see great potential in using SBC to adjust misscalibration instead of testing like Andew said. Draft building in #52.

[martinmodrak]

How big is stantarget package?

No idea :-) Just know that it exists and has some users (judging from forum discussions)

I see great potential in using SBC to adjust misscalibration instead of testing like Andew said.

I think we can definitely quantify miscalibration (I have a code that IMHO provides good approximate answer and hopefully can make it more exact), not sure if we can "adjust" for it (that sounds much harder).

[hyunjimoon]

provides good approximate answer

May I ask what you mean by this?