Given a vector of e-values \(e_1, \ldots, e_m\) (where each \(e_i\) represents the evidence against the intersection null hypothesis for model \(i\)), computes the closed-testing adjusted e-values using the arithmetic mean as the e-merging function.
Value
A numeric vector of the same length, containing the closed-testing adjusted e-values \(e^\star_i\), in the original (unsorted) order.
Details
This implements the Accelerated E-Value Calibration algorithm (Tim Stephan, ETH Zurich; Section H of the supplementary material to Arnold et al., 2026), which solves the Vovk & Wang (2021) closed-testing minimization in \(O(m \log m)\) time rather than the naive \(O(m^2)\).
Examples
raw_evalues <- c(10, 2, 1)
# Model 1 has strong evidence against it, Model 3 has none.
vovk_wang_merge(raw_evalues)
#> [1] 4.333333 1.500000 1.000000