ch.overlapToDirOv: This function converts Overlap and direction to a directional...

In ccpluncw/ccpl_R_chValues:

This function converts Overlap and direction to a directional overlap.

Description

Assume two overlapping distributions, D1 and D2. Randomly select a value from D1 and D2. p(D1 > D2) is the probability that you select a value from D1 that is greater than D2. Overlap = 1 - [abs (p(D1 > D2) - 0.5) / 0.5]. Directional Overlap = abs(overlap - 1) * direction. Directional overlap goes from -1 (no overlap D1 smaller than D2) to 0 (D1 == D2) to 1 (no overlap D1 larger than D2)

Usage

ch.overlapToDirOv(overlap, direction)

Arguments

overlap	A number that equals 1 - [abs (p(D1 > D2) - 0.5) / 0.5].
direction	A number that equals -1 or 1. The direction specifies the whether D1 or D2 was the greater distribution with the greater value. It is output when p(D1 > D2) is calculated with the bootstrap.

Value

Directional Overlap

Examples

ch.overlapToDirOv (0.5, 1)

ccpluncw/ccpl_R_chValues documentation built on Feb. 12, 2024, 4:21 a.m.