value_remaining: Potential Value Remaining

In robinvanemden/contextual: Simulation and Analysis of Contextual Multi-Armed Bandit Policies

Description

Compute "value_remaining" in arms not currently best in binomial bandits

Usage

value_remaining(x, n, alpha = 1, beta = 1, ndraws = 10000)

Arguments

x	Vector of the number of successes per arm.
n	Vector of the number of trials per arm.
alpha	Shape parameter alpha for the prior beta distribution.
beta	Shape parameter beta for the prior beta distribution.
ndraws	Number of random draws from the posterior.

Value

Value_remaining distribution; the distribution of improvement amounts that another arm might have over the current best arm.

Author(s)

Thomas Lotze and Markus Loecher

Examples

x <- c(10,20,30,80)
n <- c(100,102,120,240)
vr <- value_remaining(x, n)
hist(vr)

# "potential value" remaining in the experiment
potential_value <- quantile(vr, 0.95)

robinvanemden/contextual documentation built on Aug. 12, 2019, 9:30 p.m.