bayesplain.core.hierarchical.between_group_variance

bayesplain.core.hierarchical.between_group_variance(means, standard_errors)[source]

Estimate the spread between true group means, by DerSimonian-Laird.

Parameters:
means : array_like

Observed group means.

standard_errors : array_like

Standard error of each group mean.

Returns:

The estimate of \(\tau^2\), floored at zero. A value of exactly zero means the observed spread between groups is no larger than their individual noise can account for, in which case full pooling is the implied answer.

Return type:

float