Title :
Events with exponentially vanishing probability have exponentially growing waiting time
Author_Institution :
Information Syst. Lab., Stanford, CA, USA
fDate :
29 Jun-4 Jul 1997
Abstract :
A stationary ergodic process {Yt} with distribution Q on the sequence space 𝒴Z is examined. The probability mass Q(Yk) decays and the recurrence time ℛ(Yk) grows exponentially with rate ℋ(Q), the entropy rate. Rather than searching back for the first recurrence of the typical sequence Yk, we wait until the first occurrence of a rare event. We prove that for if Q satisfies certain mixing conditions, then there exists a polynomially growing sequence gk
Keywords :
channel capacity; entropy; estimation theory; exponential distribution; polynomials; distribution; entropy rate; exponentially growing waiting time; exponentially vanishing probability; mixing conditions; polynomially growing sequence; probability mass; rare event occurrence; recurrence time; sequence space; stationary ergodic process; Distortion measurement; Electrostatic discharge; Entropy; Information systems; Integrated circuit testing; Polynomials; Q measurement; Rate-distortion; Yttrium;
Conference_Titel :
Information Theory. 1997. Proceedings., 1997 IEEE International Symposium on
Conference_Location :
Ulm
Print_ISBN :
0-7803-3956-8
DOI :
10.1109/ISIT.1997.612952