Title :
Maximum entropy and the Edgeworth expansion
Author :
Harremoës, Peter
Author_Institution :
Dept. of Math., Copenhagen Univ., Denmark
fDate :
29 Aug.-1 Sept. 2005
Abstract :
For sums of i.i.d. random variables the maximum entropy distribution with respect to the first moments fixed is compared with the Edgeworth expansion. It is demonstrated that the Edgeworth expansion can and shall be considered as a linear extrapolation of the maximum entropy distribution. The coefficients in the Edgeworth expansion can be used as a first approximation for numerical calculation of the maximum entropy distribution.
Keywords :
extrapolation; maximum entropy methods; sequences; Edgeworth expansion; fixed first moments; linear extrapolation; maximum entropy distribution; random variables; Bismuth; Character generation; Chromium; Convergence; Entropy; Extrapolation; Gaussian distribution; Gold; Polynomials; Random variables;
Conference_Titel :
Information Theory Workshop, 2005 IEEE
Print_ISBN :
0-7803-9480-1
DOI :
10.1109/ITW.2005.1531858
