ess_rand

ess_rand(designs, y_mat, *, contrast=(1.0, -1.0))[source]

Randomization-based effective sample size of each design.

For each design, the difference-in-means estimator of the contrast is computed on every assignment, and its variance across assignments is the design’s precision.

Parameters:

  • designs (Sequence[np.ndarray]) – One array per design, each with one assignment per row.

  • y_mat (np.ndarray) – Potential outcomes, one column per group in sorted label order.

  • contrast (Sequence[float]) – Coefficient for each group.

Returns:

ESS per design; the most precise design has ESS equal to N.

Return type:

np.ndarray