Continuous MaxEnt distributions in Mathematica: a 'parameter-free' approach

Stokes, Barrie James

Title: Continuous MaxEnt distributions in Mathematica: a 'parameter-free' approach
Creator: Stokes, Barrie James
Relation: 29th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering. Bayesian Inference and Maximum Entropy Methods in Science and Engineering: 29th International Workshop on Bayesian Inference and Maximum Entropy Methods in Science and Engineering (Oxford, MI 5-10 July, 2009) p. 292-301
Publisher Link: http://dx.doi.org/10.1063/1.3275626
Publisher: American Institute of Physics
Resource Type: conference paper
Date: 2009
Description: Mathematica is used to develops a method of obtaining continuous MaxEnt distributions using the Lagrange Multiplier method that works directly in the discrete case. The continuous PDF (probability density function) is described by list of coordinates of the form {{abs₁, ord₁}, {abs₂, ord₂}, ... , {absn, ordn}}, where the abscissae absᵢ are specified numerically so as to define the domain of the PDF, and the ordinates ordᵢ are initially general symbolic variables that are assigned numeric values via the procedure to be described. In all the applications given here, n = 51. A set of Lagrange Multiplier equations is solved for the ordi, making use of the fact that the Mathematica function Interpolation[], which constructs interpolationFunction objects normally used for numeric data and for describing numerical solutions to differential equations, can operate on "semi-symbolic" data.
Subject: probability; mathematical operators; functional analysis; integral equations; entropy
Identifier: http://hdl.handle.net/1959.13/920237
Identifier: uon:9105
Identifier: ISBN:9780735407299
Language: eng
Full Text
Reviewed

Hits: 1296
Visitors: 1577
Downloads: 366

		Thumbnail	File	Description	Size	Format
View Details Download			ATTACHMENT01	Publisher version (open access)	446 KB	Adobe Acrobat PDF	View Details Download