ebScores: Construct an openEBGM object

In openEBGM: EBGM Disproportionality Scores for Adverse Event Data Mining

Construct an openEBGM object

Description

ebScores calculates EBGM scores as well as the quantiles from the posterior distribution and returns an object of class openEBGM.

Usage

ebScores(processed, hyper_estimate, quantiles = c(5, 95), digits = 2)

Arguments

processed	A data frame resulting from running processRaw.
hyper_estimate	A list resulting from running autoHyper.
quantiles	Either a numeric vector of desired quantiles to be calculated from the posterior distribution or NULL for no calculation of quantiles.
digits	A whole number scalar specifying how many decimal places to use for rounding EBGM and the quantiles scores.

Details

This function takes the processed data as well as the hyperparameter estimates and instantiates an object of class openEBGM. This object then contains additional calculations, such as the EBGM score, and the quantiles that are supplied by the quantiles parameter at the time of calling the function.

The function allows for the choice of an arbitrary amount of quantiles or no quantiles at all to be calculated. This may be helpful for large datasets, or when the EBGM score is the only metric of interest.

Value

An openEBGM object (class S3) containing:

• data: A data frame containing the results (scores, etc.).

• hyper_parameters: A list containing the hyperparameter estimation results.

• quantiles: The chosen percentiles.

Examples

data.table::setDTthreads(2)  #only needed for CRAN checks
theta_init <- data.frame(
  alpha1 = c(0.5, 1),
  beta1  = c(0.5, 1),
  alpha2 = c(2,   3),
  beta2  = c(2,   3),
  p      = c(0.1, 0.2)
)
data(caers)
proc <- processRaw(caers)
squashed <- squashData(proc, bin_size = 300, keep_pts = 10)
squashed <- squashData(squashed, count = 2, bin_size = 13, keep_pts = 10)
suppressWarnings(
  hypers <- autoHyper(data = squashed, theta_init = theta_init)
)
obj <- ebScores(processed = proc, hyper_estimate = hypers, quantiles = 5)
print(obj)

openEBGM documentation built on Sept. 15, 2023, 1:08 a.m.