extract_lrstat_matrix: Extract various summary measures from a pvlrt object

View source: R/pvlrt-postprocessing-utils.R

extract_lrstat_matrixR Documentation

Extract various summary measures from a pvlrt object

Description

Extract various summary measures from a pvlrt object

Usage

extract_lrstat_matrix(object, ...)

extract_p_value_matrix(object, ...)

extract_zi_probability(object, ...)

extract_n_matrix(object, ...)

extract_significant_pairs(object, significance_level = 0.05, ...)

extract_run_time(object, ...)

Arguments

object

a pvlrt object, which is the output of the function pvlrt or one of its wrappers such as lrt_zi_poisson, lrt_poisson and lrt_vanilla_poisson.

...

other input parameters. Currently unused.

significance_level

numeric. Level of significance.

Value

  • extract_lrstat_matrix returns the matrix of the computed log-likelihood ratio test statistics for signals. This produces a result identical to applying as.matrix.

  • extract_p_value_matrix returns the matrix of computed p-values associated with the likelihood ratio tests.

  • extract_zi_probability returns a vector of (estimated) zero-inflation probabilities.

  • extract_n_matrix returns the original contingency table (matrix) used.

  • extract_significant_pairs returns a data.table listing the AE/drug pairs determined to be significant at the provided significance level. This is essentially a subset of the data.table obtained through summary.pvlrt() that satisfies the provided significance threshold.

  • extract_run_time returns a difftime object measuring the total CPU time needed to run the original pvlrt call.

See Also

pvlrt

Examples


# 500 bootstrap iterations (nsim) in the example below
# are for quick demonstration only --
# we recommended setting nsim to 10000 (default) or bigger

test1 <- pvlrt(statin46, test_zi = TRUE, nsim = 500)
extract_lrstat_matrix(test1)
extract_p_value_matrix(test1)
extract_zi_probability(test1)
extract_n_matrix(test1)
extract_significant_pairs(test1)



pvLRT documentation built on March 7, 2023, 7:17 p.m.