A Weighted Generalized Maximum Entropy Estimator with a Data-driven Weight Academic Article

abstract

• The method of Generalized Maximum Entropy (GME), proposed in Golan, Judge and Miller (1996), is an information-theoretic approach that is robust to multicolinearity problem. It uses an objective function that is the sum of the entropies for coefficient distributions and disturbance distributions. This method can be generalized to the weighted GME (W-GME), where different weights are assigned to the two entropies in the objective function. We propose a data-driven method to select the weights in the entropy objective function. We use the least squares cross validation to derive the optimal weights. Monte Carlo simulations demonstrate that the proposed W-GME estimator is comparable to and often outperforms the conventional GME estimator, which places equal weights on the entropies of coefficient and disturbance distributions. 2009 by the author; licensee Molecular Diversity Preservation International, Basel, Switzerland.

authors

• Wu, Ximing

published proceedings

• ENTROPY

altmetric score

• 3

author list (cited authors)

• Wu, X.

citation count

• 22

complete list of authors

• Wu, Ximing

publication date

• December 2009

publisher

• MDPI  Publisher

published in

• Entropy: international and interdisciplinary journal of entropy and information studies  Journal

keywords

• Cross Validation
• Generalized Maximum Entropy Method
• Maximum Entropy

Digital Object Identifier (DOI)

• 10.3390/e11040917

URI

• https://hdl.handle.net/1969.1/182780

start page

• 917

end page

• 930

volume

• 11

issue

• 4

URL

• http://dx.doi.org/10.3390/e11040917