The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
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.
"lnorm"
.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
and
upper_tail_dist
, replaced by a single
tail_dist
that is used in both tails. We now recommend the
use of tail_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 of tail_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 than dup_tol
), and
zero_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.
q
is provided as input.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.