MATH2647 Probability II 2021
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
MATH2647
ProbabilityII
2021
Q1 A random symbol generator produces each of 10 digits and each of 26 letters
of the latin alphabet with positive probability. Assuming that the individual out- comes are independent, let M be event that an infinite sequence of such symbols contains the word MATH2647 infinitely often.
1.1 Find the probability of M using a suitable monotone approximation.
1.2 Find the probability of M using the Borel-Cantelli lemma.
In your answer you should clearly state and carefully apply every result you use.
Q2 An XOR gate adds bits according to the following rules:
0 + 0 = 1 + 1 = 0 , 1 + 0 = 0 + 1 = 1 .
Suppose that random bits (Bk )k(n)=1 are independent and have a common distribu- tion P(Bk = 1) = p = 1 P(Bk = 0), where 0 ❁ p ❁ 1. Write Sn for the result of the sum of these bits using XOR gates, and let pn be the probability of the event ❢Sn = 1❣, equivalently,
pn = P ✏the sequence (Bk )k(n)=1 contains an odd number of ‘ones’ ✑ .
2.1 Compute p1 and p2 .
2.2 Show that, with properly defined p0 , we have pn = p + (1 2p) pn 1 for all n ✕ 1.
2.3 Use generating functions to derive a closed formula for pn , and check that it gives correct values for n = 0, 1, 2.
2.4 Find nlim✦✶pn and explain your result.
Q3 Consider a Markov chain with state space ❢1, 2, 3, 4❣ and the transition matrix
✵ 0 1❂2 0 1❂2✶
P = ❇(❇)1❂2 0 1❂2 0 ❈(❈)
❈❆ .
3.1 Describe the class structure of this Markov chain and determine the period of each state.
3.2 Find all stationary distributions for this Markov chain.
3.3 Find a closed expression in terms of powers of the eigenvalues of P for the n-step transition probabilities p1(n)) and p .
3.4 Deduce closed expressions in terms of powers of the eigenvalues of P for the n-step transition probabilities p and p, where n ❃ 0.
3.5 Classify all states of this Markov chain into transient and recurrent.
In your answer you should clearly state and carefully apply every result you use.
Q4 For real an ❃ 0 and pn ✷ (0, 1), let (Xn )n✕1 be random variables such that P(Xn = an ) = pn = 1 P(Xn = 0).
For each of the following claims, prove the result if it is correct, or find a coun- terexample otherwise:
4.1 If an ✦ 0, then Xn ✦ 0 in Lr , for some r ❃ 0, as n ✦ ✶ .
4.2 If an ✦ 0, then Xn ✦ 0 in probability as n ✦ ✶ .
4.3 If an ✦ 0, then Xn ✦ 0 almost surely as n ✦ ✶ .
4.4 If pn ✦ 0, then Xn ✦ 0 in Lr , for some r ❃ 0, as n ✦ ✶ .
4.5 If pn ✦ 0, then Xn ✦ 0 in probability as n ✦ ✶ .
4.6 If pn ✦ 0, then Xn ✦ 0 almost surely as n ✦ ✶ .
4.7 If pn ✦ 0 and the variables Xn are independent, then Xn ✦ 0 in Lr , for some r ❃ 0, as n ✦ ✶ .
4.8 If pn ✦ 0 and the variables Xn are independent, then Xn ✦ 0 in probability as n ✦ ✶ .
4.9 If pn ✦ 0 and the variables Xn are independent, then Xn ✦ 0 almost surely as n ✦ ✶ .
In your answer you should clearly state and carefully apply any result you use.
2022-05-10