Rearrangement Algorithm and Maximum Entropy

Abstract

We study properties of the Rearrangement Algorithm (RA) in the context of inferring dependence among variables given their marginal distributions and the distribution of the sum. We show that the RA yields solutions that are “close to each other” and exhibit almost maximum entropy. The consequence of this result is that the RA can be used as a stable algorithm for inferring a dependence that is economically meaningful.

Keywords: Entropy, Block Rearrangement Algorithm, Inferring dependence, Maximum determinant

JEL Classification: C60