find_intercept_iter: Find the Intercept for a Dataset

In bips-hb/rgp: Identification of Risk Groups in Pharmacovigilance Using Penalized Regression and Machine Learning (RGP)

Description

This function approximates intercept (β_{0}) iteratively to get a balanced case-control ratio in the simulated data. This function uses simstudy package.

Usage

find_intercept_iter(true_beta, dtx_old, ccr = 0.5, ...)

Arguments

true_beta	A vector of true beta values including both assciated and non-associated variables.
dtx_old	Data.table of old dataset to be appended with outcome.
ccr	Numeric, case-control ratio (Default: 0.5).
cepts	A numeric list of intercepts to use, recommmended to be NULL (Default: NULL).
iter	Integer, number of increments between intercept limits (Default: 500).

Value

A list of

• data: The new data.table with outcome column data$out.

• intercept: An estimation of the intercept.

• difference: The absolute difference between ccr and mean(data$out).

• sequence: Sequence of tested intercepts.

• ccr: Case-to-control ratio for each intercept.

See Also

defDataAdd and addColumns in simstudy package.

Examples

dtx <- simstudy::genCorGen(n = 100, nvars = 5, params1 = 0.15,
 dist = "binary", wide = TRUE, rho = 0.1, corstr = "ar1")
find_intercept_iter(c(-2, -2, 2, 2, 2), dtx)

bips-hb/rgp documentation built on Feb. 3, 2021, 11:31 a.m.