Title :
Asymptotically optimal function estimation by minimum complexity criteria
Author :
Barron, Andrew ; Yang, Yuhong ; Yu, Bin
Author_Institution :
Dept. of Stat., Yale Univ., New Haven, CT, USA
fDate :
27 Jun-1 Jul 1994
Abstract :
The minimum description length principle applied to function estimation can yield a criterion of the form log(likelihood)+const·m instead of the familiar log(likelihood)+(m/2) log n where m is the number of parameters and n is the sample size. The improved criterion yields minimax optimal rates for redundancy and statistical risk. The analysis suggests an information-theoretic reconciliation of criteria proposed by Rissanen (1983), Schwarz (1978), and Akaike (1973)
Keywords :
computational complexity; information theory; minimax techniques; parameter estimation; statistical analysis; asymptotically optimal function estimation; code; information theory; minimax optimal rates; minimum complexity criteria; minimum description length; parameter estimation; redundancy; sample size; statistical risk; Estimation error; Information analysis; Length measurement; Minimax techniques; Polynomials; Quantization; Redundancy; Statistics; Stochastic processes; Yield estimation;
Conference_Titel :
Information Theory, 1994. Proceedings., 1994 IEEE International Symposium on
Conference_Location :
Trondheim
Print_ISBN :
0-7803-2015-8
DOI :
10.1109/ISIT.1994.394933
