Poole and Rosenthal's APRE pools classification errors over roll calls,
\(\sum_j (m_j - e_j) / \sum_j m_j\), which is the minority-weighted mean
of the per-vote PREs. Without minority sizes every roll call is weighted
equally. Identical to the Python arm morie.fn.apres.apre_statistic.
Usage
apre_statistic(all_pre, minority = NULL)
Arguments
- all_pre
Numeric vector of per-roll-call PREs.
- minority
Optional positive minority-side sizes, one per roll call.
Value
A list with value (APRE), mean_pre,
median_pre, min_pre, max_pre, n_roll_calls
and weighted.
References
Poole, K. T. and Rosenthal, H. (1997). Congress: A
Political-Economic History of Roll Call Voting. Oxford University Press.
Examples
apre_statistic(c(0.8, 0.5), minority = c(40, 10))$value
#> [1] 0.74
