BBUM_loglik: Log-likelihood for the BBUM model

In wyppeter/bbum: BBUM correction for significance testing of p values

Log-likelihood for the BBUM model

Description

BBUM_loglik computes the total log-likelihood of the given BBUM parameters, given data points with and without the primary beta component. This is the maximization target for fitting the BBUM model to data (MLE).

Usage

BBUM_loglik(params, posSet, negSet, limits = list(), rcap, pBBUM.alpha)

Arguments

params	Named vector of BBUM parameters.
posSet	Vector of values following a BBUM distribution including the primary beta component ("signal set" or "sample set").
negSet	Vector of values following a BBUM distribution without the primary beta component, i.e. a BUM distribution with the secondary beta distribution ("background set").
limits	Named list of custom limits for specific paramters. Parameters not mentioned would be default values.
rcap	Whether the parameter r should have a stringent upper bound in this instance (for smart toggling of outlier detection).
pBBUM.alpha	Cutoff level of BBUM-FDR-adjusted p values for significance testing. Only used here to generate appropriate default limits.

Details

p values lower than .Machine$double.xmin*10 are constrained to that limit value before evaluation to avoid zero and machine limit issues.

Value

The value of the total log-likelihood. The logarithm used is the natural logarithm.

wyppeter/bbum documentation built on Oct. 3, 2023, 3:29 p.m.