final_sol_2008

final_sol_2008 - MS&E 221 Final Examination SOLUTIONS...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: MS&E 221 Final Examination - SOLUTIONS Ramesh Johari March 20, 2008 Part A (4 questions; 5 points each; 20 points total) Answer each of the following short answer questions. Explain your answers! (?? points per question) 1. True or false : Every Markov chain has at least one closed class. (Explain if true, and provide a counterexample if false.) Answer: FALSE. See Figure 1 for a counterexample. Note that it does not suffice to give an example of an infinite chain that is transient – remember that for infinite communicating classes, a class can be transient and still be closed! 2 points for correct answer; 3 points for correct justification. 2. You receive two types of e-mail: spam, and mail from friends. Spam arrives according to a Poisson process of rate α S , and mail from friends as a Poisson process of rate α F . There is a probability p that a piece of spam is caught by your spam filter, and never enters your inbox. Find the probability that at least 3 spam messages enter your inbox before the first message from a friend. Answer: Spam enters the inbox as a Poisson process of rate (1- p ) α S . So the answer is: parenleftbigg (1- p ) α S (1- p ) α S + α F parenrightbigg 3 1 point for model; -1 point for minor math mistake. 3. True or false : Every Markov chain has at least one transient state. (Explain if true, and provide a counterexample if false.) Answer: FALSE. Any irreducible, recurrent chain is a counterexample. 2 points for correct answer; 3 points for correct justification. 4. Consider an irreducible discrete time Markov chain on the state space { 1 , 2 ,...,N } , with transition matrix P . Suppose that all column sums are equal to one: i.e., for all j , ∑ i P ij = 1 . Find the unique invariant distribution of the resulting Markov chain. Is any such Markov chain reversible? Answer: The unique invariant distribution is the uniform distribution, π ( i ) = 1 /N for all i . For such a chain to be reversible (or more precisely, for there to exist a two-sided stationary version that is reversible), the transition matrix must be symmetric, i.e., P ij = P ji for all i,j . 2 points for writing basic invariant dist. equations; 1 point for solving; and 2 points for reversibility explanation. 1 Part B (3 questions; 80 points total) Problem 1. Consider a discrete time Markov chain on states 1 , 2 , 3 , 4 , 5 , 6 with the following transition matrix: P = 1 / 2 1 / 2 1 / 2 1 / 2 1 / 2 1 / 2 1 / 6 1 / 6 1 / 6 1 / 6 1 / 6 1 / 6 1 / 2 1 / 2 1 / 2 1 / 2 (a) (5 points) Find the communicating classes. Which are transient? recurrent? closed?...
View Full Document

This note was uploaded on 02/03/2011 for the course MS&E 221 taught by Professor Ramesh during the Winter '11 term at Stanford.

Page1 / 7

final_sol_2008 - MS&E 221 Final Examination SOLUTIONS...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online