This is a tool to find the optimal rerandomization threshold in non-sequential experiments. We offer three procedures based
on assumptions made on the residuals distribution: (1) normality assumed (2) excess kurtosis assumed (3) entire distribution assumed.
Illustrations are included. Also included is a routine to unbiasedly estimate Frobenius norms of variance-covariance matrices.
Details of the method can be found in "Optimal Rerandomization via a Criterion that Provides Insurance Against Failed Experiments"
Adam Kapelner, Abba M. Krieger, Michael Sklar and David Azriel (2020)