Density for bounded distributions

[drcthal]

This is more of a general kdensity issue, but if you feed in a bounded distribution, the distribution plot will show/color area outside of the bounds, while the bar plot will (correctly) not reflect that.
e.g.
set.seed(123)
prior <- runif(100)
IMPosterior(prior = prior)

[ignacio82]

Good catch. The only solution I can think is adding an argument "bounded" that you can set to TRUE and if you do that we can avoid plotting things outside the domain of the draws that are passed. Do you think that would work? do you have a better idea?

[ignacio82]

On seconds thoughts maybe the argument that we want to add is support and just pass that to kdensity

[drcthal]

Hmm, functionally it looks like calculating densities with kdensity and support doesn't look too different from just calculating regular densities and chopping them off, so for simplicity (not changing the code and adding the dependency) I think sticking with density and chopping the ends off makes sense. It'd still be good to call the option support for consistency.

[ignacio82]

ok