ideal_pcc: Determine the Ideal PCC

Description Usage Arguments Value Author(s) References Examples

View source: R/ideal_pcc.R

Description

Determine the probability of correct classification (PCC) for a study employing the ideal classifier. The ideal classifier is constructed assuming we know exactly the important features and their effect size. The ideal PCC is the uppper bound of the PCC of any linear classifiers.

Usage

1
ideal_pcc(mu0, m, p1 = 0.5)

Arguments

mu0

The effect size of the important features.

m

The number of the important features.

p1

The prevalence of the group 1 in the population, default to 0.5.

Value

The PCC of the ideal classifier.

Author(s)

Meihua Wu <[email protected]> Brisa N. Sanchez <[email protected]> Peter X.K. Song <[email protected]> Raymond Luu <[email protected]> Wen Wang <[email protected]>

References

Dobbin, Kevin K., and Richard M. Simon. 2007. "Sample Size Planning for Developing Classifiers Using High-dimensional DNA Microarray Data." Biostatistics 8 (1) (January 1): 101-117.

Examples

1
2
ideal_pcc(mu0=0.4, m=10, p1 = 0.6) 
#return: 0.8999055

Example output

[1] 0.8999055

HDDesign documentation built on May 29, 2017, 8:13 p.m.