Title :
Optimal Bregman prediction and Jensen´s equality
Author :
Banerjee, Arindam ; Guo, Xin ; Wang, Hui
Author_Institution :
Dept. of Electr. & Comput. Eng., Texas Univ., Austin, TX, USA
fDate :
27 June-2 July 2004
Abstract :
This paper provides necessary and sufficient conditions for general loss functions under which the conditional expectation is the unique optimal predictor of a random variable. Further, using such loss functions, we give an exact characterization of the difference between the two sides of Jensen´s inequality.
Keywords :
optimisation; prediction theory; random processes; stochastic processes; Jensen´s inequality; optimal Bregman prediction; random variable; Euclidean distance; Least squares methods; Random variables; Sufficient conditions;
Conference_Titel :
Information Theory, 2004. ISIT 2004. Proceedings. International Symposium on
Print_ISBN :
0-7803-8280-3
DOI :
10.1109/ISIT.2004.1365205
