powered by
Compute "value_remaining" in arms not currently best in binomial bandits
value_remaining(x, n, alpha = 1, beta = 1, ndraws = 10000)
Vector of the number of successes per arm.
Vector of the number of trials per arm.
Shape parameter alpha for the prior beta distribution.
Shape parameter beta for the prior beta distribution.
Number of random draws from the posterior.
Value_remaining distribution; the distribution of improvement amounts that another arm might have over the current best arm.
# NOT RUN { 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) # }
Run the code above in your browser using DataLab