Title: A linearisation of the maximum entropy formalism using separable programming

Authors: Ermanno Affuso; Steven B. Caudill

Addresses: Department of Economics and Finance, Mitchell College of Business, University of South Alabama, 5811 USA Drive S., Mobile, AL 36688, USA ' Department of Economics, Rhodes College, 2000 N Parkway, Memphis, TN 38112, USA

Abstract: The maximum entropy principle is a standard tool for the calibration of non-linear programming models which are frequently used for policy analysis. The information entropy function is concave and separable. In this paper, we derive a linear approximation of the entropy using separable programming. As we demonstrate, our linear entropy formulation is useful for the calibration of separable non-linear models of very large scale. To demonstrate, we solve both an ill-posed and a well-posed inverse problem and we analyse the sensitivity of the results on the number of breakpoints of the piecewise linear approximation.

Keywords: maximum entropy; entropy econometrics; separable programming; concave programming; linear approximation; linearisation; model calibration; nonlinear programming; nonlinear models; policy analysis; modelling.

DOI: 10.1504/IJOR.2015.068558

International Journal of Operational Research, 2015 Vol.22 No.4, pp.385 - 404

Received: 21 Sep 2012
Accepted: 14 Jun 2013
Published online: 09 May 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article