Conditional entropy. Let A and B = (Bo, B₁,..., Bn) be a discrete random variable and vector, respectively. The conditional entropy of A with respect to B is defined as H(A | B) = E (E{-log f(A | B) | B}) where f(a | b) = P(A = a | B = b). Let X be an aperiodic Markov chain on a finite state space. Show that and that H(Xn+1 | X0, X₁. , Xn) = H(Xn+1 | Xn), .... H(Xn+1 | Xn) →→лi Σ Pij i j if X is aperiodic with a unique stationary distribution . log Pij as n →∞,

Conditional entropy. Let A and B = (Bo, B₁,..., Bn) be a discrete random variable and vector, respectively. The conditional entropy of A with respect to B is defined as H(A | B) = E (E{-log f(A | B) | B}) where f(a | b) = P(A = a | B = b). Let X be an aperiodic Markov chain on a finite state space. Show that and that H(Xn+1 | X0, X₁. , Xn) = H(Xn+1 | Xn), .... H(Xn+1 | Xn) →→лi Σ Pij i j if X is aperiodic with a unique stationary distribution . log Pij as n →∞,

Elementary Linear Algebra (MindTap Course List)

8th Edition

ISBN:9781305658004

Author:Ron Larson

Publisher:Ron Larson

Chapter2: Matrices

Section2.5: Markov Chain

Problem 55E

See similar textbooks