Posterior probability P(p2 > p1 | data) under independent Beta priors

Uses Beta posteriors induced by the analysis priors to compute \(P(p_2 > p_1 \mid y_1, y_2)\).

Usage

postProbHplus(y1, y2, n1, n2, a_1_a = 1, b_1_a = 1, a_2_a = 1, b_2_a = 1)

Arguments

y1

Number of successes in arm 1 (control).

y2

Number of successes in arm 2 (treatment).

n1

Sample size in arm 1.

n2

Sample size in arm 2.

a_1_a, b_1_a

Shape parameters of the Beta prior for \(p_1\) under the alternative (analysis prior).

a_2_a, b_2_a

Shape parameters of the Beta prior for \(p_2\) under the alternative (analysis prior).

Value

Numeric scalar, posterior probability \(P(p_2 > p_1 | y_1, y_2)\).