The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
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[@]dogado.de.
power_best_binomial()sim_power_best_binomial()power_best_binomial() produces a
valid result using simulations.wcs_power_best_binomial() to
avoid search probabilities lower than the difference.sim_power_nbinom() to explicitly indicate the
direction of the alternative hypothesis as it could be wrongly inferred
from the rr alone.multz() and multq() to follow R
convention in distributions. A new parameter lower.tail is
addedss_best_normal() returns now an integer
valuess_ni_ve() now computes the total number of events
using Freedman’s (1982) log-rank sample size formula instead of
gsDesign::nBinomial1Sample(), which better reproduces Table
1 of Fleming et al. (2021) and removes the gsDesign package
dependencyve_exp/ve_ac parameters to
ss_ni_ve(), allowing the experimental vaccine’s assumed
efficacy to differ from the active comparator’s, which reproduces Table
2 of Fleming et al. (2021)true_diff parameter to
sim_power_ni_normal() so simulations are no longer
restricted to a true difference of 0 between groups, allowing
sensitivity evaluation when the experimental group is truly slightly
worse or better than controlprophr() to accept vector inputs for
p0/hr without an “condition has length > 1”
errorsim_power_best_bin_rank()
selects the group with the highest total rank as best (roxygen
said “lowest”); power_best_normal() returns a probability,
not an integer sample size; sim_power_ni_normal()’s and
sim_power_equivalence_normal()’s t_level
examples/docs clarifiedsim_power_nbinom()’s examplepower_events_rate() function to compute the
exact binomial probability of observing at least a given number of
events, across combinations of sample sizes and risks.power_events_rate() alongside
power_single_rate().sim_best_binomial() function to select the best
using ranks rather than the custom approach of which == max().sim_.ni_fleming has been renamed to ni_ve.multz function was modified to correctly handle the
case when there is only one group.probhr function to estimate the probability
of an event in the experimental group based on the control group
probability and the hazard ratio, assuming proportional hazards.ni_fleming function.ni_fleming.power_best_binomial and power_best_normal
were updated to match their rank-based equivalents.dif represents the difference between the most
promising group and the rest.prob is now the probability in the
most promising group (previously it referred to the other groups), and
dif defines how much lower the other groups are.power_best_norm_ranks and
power_best_bin_rank.power_best_norm_rank to empirically estimate
power for normal distributions based on ranks.power_test_binomial where, in the
absence of ties, the sample function randomly selected a
number instead of sampling from a range.power_best_normal.power_best_binomial now returns a data frame with power
and 95% confidence interval.power_best_normal now returns a data frame with power
and 95% confidence interval.power_ni_normal now includes the number of simulations
in the output data frame.power_best_bin_rank, which selects the best group
based on ranks, instead of assuming the highest value as in
power_best_binomial.power_ni_normal function.ggplot_prob_lowest_power
graph.power_equivalence_normal function.power_best_binomial function and related
helpers.power_best_normal function and related
helpers.NEWS.md file to track package changes.power_single_rate function.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.
Health stats visible at Monitor.