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.
"control"
when control_augment = 0 in
mab_from_rct()multiple_mab_simulation() and
single_mab_simulation() are removed in favor of
mab_from_rct() which performs the same functions. Multiple
simulations specified via the r argument.mab_from_rct() now accepts bare column names and no
longer accepts stringsmultiple.mab and mab classes have been
removed in favor of multi_rct_mab,
multi_param_mab, single_rct_mab,
single_param_mab, and .mab.plot(), summary(), and
print() methods have been removed pending new
versions.joint_test() a function accepting single_*_mab
classes. Methods are "bootstrap" and
"randomization" inspired by Offer-Westort et. al (2021),
see docs for more.simulate_mab() is a new function that simulates a
Multi-Arm-Bandit trial from provided population parameters. Uses similar
options to mab_from_rct(). See the
simulate_mab() documentation for details.simulate_mab()
and mab_from_rct() can now produce linear contrasts
(e.g. treatment effects) in final output along with traditional mean
estimates for all estimators. See the documentation for which options
are available.simulate_mab() and mab_from_rct()
documentation for more info.random_assign_prop now adjusts the assignment
probabilities instead of splitting the data; see docs for more
details.tanf dataset with individual assignment from 20-30
seconds on v.0.3.0 to 9 seconds.data.table versus
tibble data.frame classes. Ensuring results are data.frame
library agnostic. Fixing miscellaneous bugs for data.table
inputs.plot.mab() for type = "assign" now
displays the proportion of total observations assigned to each treatment
for each period, instead of the individual probability of
assignment.multiple_mab_simulation() calculates the number of
observations assigned to each treatment for each trial, and provides
support for plotting them has been added to
plot.multiple.mab() via the type = "hist" and
quantity = "assignment" arguments.summary.mab() now includes a new column with the number
of observations assigned to each treatment.summary.multiple.mab() now includes two new columns
with the mean and standard deviation for the number of observations
assigned to each treatment across the simulations.time_unit = "month" can now be
specified with and without an appropriate month_col,
resulting in either time-based (no month_col) or
calendar-based (provided month_col) assignments. See the
time_unit documentation for more details.plot.multiple.mab() now accepts arguments for
ggplot2::facet_grid() for more precise customizations.condition_col of the data_cols argument.single_mab_simulation() or
multiple_mab_simulation(), with the same seeds on the same
system, now results in the same outcome regardless of input data
class.multiple_mab_simulation() supports parallel processing
via future.single_mab_simulation() and
multiple_mab_simulation() support data.table for
larger data sets.summary(), print(), and
plot() generics for mab and
multiple.mab class objects.single_mab_simulation() and
multiple_mab_simulation() throw informative error messages,
relating to argument specification, and data types passed.multiple_mab_simulation().single_mab_simulation() and
multiple_mab_simulation() simulate successfully.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.