On some properties of the NML estimator for Bernoulli strings

Authors:
Ciprian Doru Giurcăneanu
Affiliations:
Institute of Signal Processing, Tampere University of Technology P.O. Box 553, FIN-33101 Tampere, Finland
Venue:
Information Processing Letters
Year:
2004

Citing 6
Cited 0

Stochastic Complexity in Statistical Inquiry Theory

Stochastic Complexity in Statistical Inquiry Theory
On asymptotics of certain sums arising in coding theory

IEEE Transactions on Information Theory - Part 2
Fisher information and stochastic complexity

IEEE Transactions on Information Theory
The minimum description length principle in coding and modeling

IEEE Transactions on Information Theory
Strong optimality of the normalized ML models as universal codes and information in data

IEEE Transactions on Information Theory
Bounds on the worst case probability of undetected error

IEEE Transactions on Information Theory

Quantified Score

Hi-index	0.89

Visualization

Abstract

This paper proves the monotonicity of the sequence Cn/√n, where Cn denotes the normalization coefficient in the universal Normalized Maximum Likelihood (NML) model for the Bernoulli class. The main result is used to find a nonasymptotic estimation of log Cn.