Changelog
distfromq 1.0.4
- Fix a bug where floating point issues could cause adjusted probability levels to fall outside the interval [0, 1] with a point mass at 0.
distfromq 1.0.2
Fix a bug that came up with multiple discrete point masses, where floating point issues could cause adjusted probability levels to fall outside the interval [0, 1].
Add argument checking.
distfromq 1.0.1
- Fix a bug when all quantiles occur at one of three duplicated values, one of which is zero, when the tail distribution is
"lnorm"
.
distfromq 1.0.0
Major Changes
Refactored handling of discrete point masses in the distribution, with a goal of better supporting use of a lognormal distribution in the tails.
Removed separate arguments for
lower_tail_dist
andupper_tail_dist
, replaced by a singletail_dist
that is used in both tails. We now recommend the use oftail_dist = "lnorm"
when the target variable is bounded above zero. This choice will automatically handle the non-negativity constraint correctly, and will have a longer right tail than the default choice oftail_dist = "norm"
.Introduced parameters
dup_tol
, which is used for detecting point masses when quantile values are approximate duplicates (i.e., the distance between them is less thandup_tol
), andzero_tol
, which is used for detecting point masses at 0 when a lognormal family is used.Updated the vignette
Added a more comprehensive suite of unit tests.