Correction to the storage in PWD_outlier function, leading to
slightly different z-scores in the output.
Output to PWD_outlier function was adjusted to include a dataframe
element with more easily-readable output.
The option to print (via the printem argument) during the function
application was thus removed for simplicity.
Updated the primary reference to the now-known publication
date.
ppwdeming 2.0.0
Sizable update, refining functions to operate more efficiently.
PWD_RL is no longer called by other functions, but it is retained
for individual application.
PWD_get_gh is the major update, in which the original search
algorithm is replaced by an improved algorithm, which we refer to as the
unified approach, focusing on the ratio rho=sigma/kappa. This results in
a many-fold decrease in computing speed.
The interconnected functions now also incorporate an optional
starting point for all the parameters in the search algorithm:
PWD_get_gh, PWD_inference, and PWD_outlier include options to initialize
any of rho, alpha, beta, or the vector mu.
PWD_known is also updated to reflect the full likelihood, leading to
an improved algorithm using the optim function.
Various housekeeping and grammatical edits were also
integrated:
Adjusted various functions to better handle missing values.
Expand the reference to our paper by adding its title wherever we
just have the DOI.
Under PWD_get_gh Details, removed “reduces the number of optimized
parameters…”
Under PWD_inference Description, deleted “Currently”.
Under WD_Linnet Details, added “since the underlying precision
profile models are not the same.”
Removed double-spacing in message-outputs (artifact of retaining “”
from cat, no longer needed)
Inside PWD_outlier function, removed extraneous text output and
double-spacing.
Under PWD_outlier values, edited the description of
scalr.
Clarification in details for PW_outlier.
Clarification in the description of PWD_resi.
Streamlined the package Description.
Standardized the argument definitions using the parameter notation
(versus words) and ended each with a period (as is standard) instead of
a comma.
Revised the Details section for PWD_get_gh to reflect the fewer
number of parameters.
Updated the Reference for the primary reference paper with the new
doi and accepted status.
The main change was updates to the PWD_resi function: removed Pvals,
and edited documentation to match, reflecting removal from the submitted
paper.
Minor documentation edits, standardizing the annotation.
Adjusted output name to be L instead of like for -2 log
Likelihood.
Added details for default initial values of parameters.
Set printem defaults to reduce automatic output for
PWD_outlier.
Minor adjustments to documentation language.
Added Argument notes and Details for PWD_get_gh documentation,
defining rho as the ratio sigma/kappa.
Fixed output formatting for PWD_outlier.
ppwdeming 1.0.6
revised individual function documentation to reference arXiv
document by arXiv DOI in format doi:10.48550/arXiv.YYMM.NNNNN.
adjusted one other doi reference correspondingly.
ppwdeming 1.0.5
updated algorithm in function PWD_RL to better handle
rare non-convergence.
ppwdeming 1.0.4
revised primary reference in DESCRIPTION file to include author
names and to have consistent format.
For runnable (but >10s running) code in examples, replaced with .
This was applied to functions: PWD_inference, PWD_outlier.
Removed the default console output by:
setting the printem to default to FALSE; and
replacing any print()/cat() with message(). This was applied for:
PWD_get_gh, PWD_inference, PWD_known, PWD_outlier, PWD_resi, and
WD_Linnet. Relevant items are elements on the list-value of the
function.
Replaced print() with message() and stop() for possible error in
WD_General.
ppwdeming 1.0.3
placed examples directly into function-building files;
used to keep runtime under 10s.
ppwdeming 1.0.2
Corrected Github address.
ppwdeming 1.0.1
Updated arXiv URL and Description to more closely meet
auto-check.