A univariate halting test using the Anderson-Darling test.
Calculates p-value using specified halting test.
.apply_crit_to_condition_pairs
Returns smallest value from .apply_crit for all condition pairs.
Compares outputs of ldamatch runs.
A univariate halting test using the Wilcoxon test, which must be satisfied for all condition pairs.
Calculates basic metrics about ldamatch search result.
Creates halting test from multiple tests.
Criterion function for ad_halt.
Criterion function for U_halt.
Returns smallest halting_test-threshold ratio, or 0 if less than 1.
.choose_most_frequently_chosen_subject_from_subject_tuples
Chooses best one(s) of a set of subjects having the best p-value(s).
Combines current best and candidate sets, keeping the highest metric value.
Returns halting tests for names, or checks if pass functions are suitable.
Recycles threshold values for halting tests.
.choose_subject_with_best_p_value_from_subject_tuples
Chooses best one(s) of a set of subjects having the best p-value(s).
.normalize_max_removed_per_cond
Normalizes max_removed_per_cond parameter for match_groups() and estimate_exhaustive().
.check_subspaces_for_group_size_setup
Searches over all possible subspaces for specified group size setup.
.calc_subject_balance_divergence
Characterizes closeness of actual group sizes to what is expected.
Normalizes the props parameter for match_groups().
Returns human readable format for number of seconds.
Determines which arguments for a function, which is its caller by default.
nondeterministic_matching_methods
The available nondeterministic methods for matching.
.internally_compare_ldamatch_outputs
Compares outputs of ldamatch runs using internally normalized parameters.
Creates string from list of vectors.
Criterion function for f_halt.
An infinitesimally small amount, used to check if values are
approximately the same.
Creates all group sizes by reducing one group in all rows of grpsizes.
.create_Cartesian_iterable
Creates Cartesian product of iterators.
Uniquifies a list.
Creates a matched group via backward selection.
.search_heuristic_with_lookahead
Finds matching using depth-first search, looking ahead n steps.
Warns about extra (i.e. unused) parameters.
parallelized_matching_methods
The available parallelized methods for matching.
A univariate halting test using the t-test, which must be satisfied for all condition pairs.
.choose_best_test_statistic
Chooses rows with best test statistic.
A multivariate halting test appropriate for more than two condition levels.
Chooses best set of subjects in a set.
Gets value for ldamatch global parameter.
Calculates p-value-threshold ratio.
Criterion function for l_halt.
Calculates multipliers used in search_random.
Criterion function for ks_halt.
A univariate halting test using the Kolmogorov-Smirnov Test, which must be satisfied for all condition pairs.
Estimates the maximum number of cases to be checked during exhaustive search.
Wrapper to foreach::foreach called from .choose_best_subjects.
Flips logical vector at specified indices
Orders rows by similarity to expected group size proportions.
The available methods for matching.
Criterion function for t_halt.
A univariate halting test using Levene's test.
Searches the space backwards, prefering more subjects and certain group size
proportions.
Finds matching using depth-first search, looking ahead n steps.
A univariate halting test using Fisher's exact test.
OBSOLETE: Finds matching using depth-first search recursively.
ldamatch: Selection of Statistically Similar Research Groups.
Finds matching using depth-first search, looking ahead n steps.
Searches by randomly selecting subspaces with decreasing expected size.
Sets value for ldamatch global parameter.