Title :
On a relation between the principle of minimum relative entropy and maximum likelihood estimation
Author :
Tzannes, M.A. ; Noonan, J.P.
Author_Institution :
Dept. of Electr. Eng., Tufts Univ., Medford, MA, USA
Abstract :
A justification for the use of the minimum relative entropy (MRE) principle as a probability density function (PDF) estimation method is given by showing a relation with the classical maximum likelihood estimation procedure. This ultimately provides a noninformation theoretic argument for the validity of the MRE principle as well as some properties that the MRE PDF obeys
Keywords :
entropy; estimation theory; minimisation; probability; signal processing; maximum likelihood estimation; minimum relative entropy; probability density function; Arithmetic; Constraint theory; Digital signal processing; Entropy; Estimation theory; Maximum likelihood estimation; Minimization methods; Parameter estimation; Probability density function; Signal processing algorithms;
Conference_Titel :
Circuits and Systems, 1990., IEEE International Symposium on
Conference_Location :
New Orleans, LA
DOI :
10.1109/ISCAS.1990.112235
