nprob¶

Syntax¶

nprob(q,n)

Parameters¶

q: The total probability of success. Valid values are between 0 and 1.
n: An integer representing the number of repeated independent choices.

Returns¶

\(1-(1-q)^{(1/n)}\)

Description¶

A convenience action for computing the probability p with which each of (at most) n individual, independently-repeated trials should succeed given the cumulative probability of success q.

Equivalently, computes the success probability p for a Binomial distribution with parameters n and p (binom(n, p)) given the probability that the outcome of a draw from that distribution is at least 1.

The value p is computed as: \(1-(1-q)^{(1/n)}\)

Note that it is incorrect to break up a probability over several repeated decisions using q/n.

state A {
    wait(24)
    next(B) with prob(0.5/10)
    default(A)
}

This is wrong as it does not account for the decreasing number of agents that pass by the next statement with each repitition.

Examples¶

Suppose you want to transition between state A and state B with a total probability 0.5, but it involves repeated decisions over 10 days.

state A {
    wait(24)
    next(B) with prob(nprob(0.5, 10))
    default(A)
}