MTH6141 Random Processes, Spring 2012, Exercise Sheet 6

Please drop your answer to Question 2 in the red box on the second oor of the Maths building by 13:00 on Tuesday 6th March 2012. You are strongly encouraged to attempt at least Questions 1 and 2. Please send comments and corrections to m.jerrum@qmul.ac.uk. 1. (a) Some revision of Probability Models. Prove the following: If A has a Po() and B has a Po() distribution and A and B are independent random variables, then A + B has a Po( + ) distribution. (b) Recall Lemma 2.2 from the notes: Let X (t) and Y (t) be independent Poisson processes with rates and . The stochastic process Z (t) = X (t) + Y (t) is a Poisson process with rate + . We proved this lemma using the innitesimal description of the Poisson process. Now prove Lemma 2.2 again, but this time using the axiomatic description (Denition 2.1). (c) At Stepney Green Underground station the arrival of Hammersmith and City Line trains forms a Poisson process of rate 6 per hour and the arrival of District Line trains forms a Poisson process of rate 14 per hour. These processes are independent. What can you say about the distribution of arrivals of trains (of both kinds) at Stepney Green station?

Please see over. 1

1 per 2. Customers enter a shop according to a Poisson process with rate 5 minute. Let C (t) be the number of customers who have entered the shop after it has been open for t minutes.

(a) Calculate the following: i. ii. iii. iv. v. P(C (20) = 3), P(C (20) = 3 | C (10) = 1), P(C (10) = 1, C (20) = 3), P(C (20) = 1 | C (10) = 3), P(C (10) = 1 | C (20) = 3).

(Your answer should be left involving powers of e but simplied in all other ways. Briey explain your answers with reference to the properties of the Poisson process.) (b) What is the probability that the second customer arrives within the rst 15 minutes. (Again, simplify, but leave powers of e.) (c) Suppose that each customer spends exactly 10 minutes in the shop. Find the distribution of the number of customers in the shop after it has been open for 1 hour. Explain your reasoning. (d) Divide an hour into 30 even minutes 0, 2, 4, . . . , 58 and 30 odd minutes 1, 3, 5, . . . , 59. Find the distribution of the number of customers who arrive during the even minutes on a given hour. Hint. Recall Q1(a). 3 . (a) Recall Lemma 2.3 from the notes: Let X (t) be a Poisson processes with rate . The stochastic process X (t) dened by the thinning procedure [in which each event is retained with probability p and rejected with probability 1 p] is a Poisson process with rate p. We proved this Lemma using the innitesimal description of the Poisson process. Now prove Lemma 2.3 again, but this time using the axiomatic description (Denition 2.1). (b) At Stepney Green Underground station the arrival of trains forms a 3 a train is a Poisson process of rate 20 per hour. With probability 10 7 Hammersmith and City Line train and with 10 a District Line train, independent of the other arrivals. What can you say about the distribution of arrivals of Hammersmith and City Line trains at Stepney Green station?