English

A Matlab Program to Calculate the Maximum Entropy Distributions

Data Analysis, Statistics and Probability 2007-05-23 v1

Abstract

The classical Maximum Entropy (ME) problem consists of determining a probability distribution function (pdf) from a finite set of expectations of known functions. The solution depends on N+1N+1 Lagrange multipliers which are determined by solving the set of nonlinear equations formed by the NN data constraints and the normalization constraint. In this short communication we give three Matlab programs to calculate these Lagrange multipliers. The first considers the general case where the functions can be any functions. The second considers the special case of power functions xnx^n. In this case the data are the geometrical moments of p(x)p(x). The third considers the special case of Fourier series functions exp(jnωx)\exp(-j n \omega x). In this case the data are the trigonometrical moments of p(x)p(x). Some examples are also given to illustrate the usefullness of these programs.

Keywords

Cite

@article{arxiv.physics/0111126,
  title  = {A Matlab Program to Calculate the Maximum Entropy Distributions},
  author = {A. Mohammad-Djafari},
  journal= {arXiv preprint arXiv:physics/0111126},
  year   = {2007}
}

Comments

Presented at MaxEnt91. Appeared in Maximum Entropy and Bayesian Methods, C.R. Smith, G.J. Erickson and Paul O. Neudorfer (Ed.), pp: 221-234, Kluwer Academic Publishers (http://www.wkap.nl/prod/b/0-7923-2031-X)