This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: CS 70 Discrete Mathematics and Probability Theory Fall 2010 Tse/Wagner Soln 13 1. (25 pts.) Probabilistically Buying Probability Books Chuck will go shopping for probability books for K hours. Here, K is a random variable and is equally likely to be 1, 2, or 3. The number of books N that he buys is random and depends on how long he shops. We are told that Pr [ N = n  K = k ] = c k , for n = 1 ,..., k for some constant c . (a) Compute c . Answer: For any k , we know that probabilities conditioned on K = k must sum to 1, i.e ∑ n Pr [ N = n  K = k ] = 1 , so it must be that 1 = k ∑ n = 1 Pr [ N = n  K = k ] = k × c k = c . Thus, c = 1. (b) Find the joint distribution of K and N . Answer: The joint distribution specifies Pr [ N = n ∩ K = k ] for all n and k . Note that Pr [ N = n ∩ K = k ] = Pr [ N = n  K = k ] Pr [ K = k ] and we know Pr [ N = n  K = k ] and Pr [ K = k ] (it says all k ∈ { 1 , 2 , 3 } are equally likely). We use this formula to calculate Pr [ N = n ∩ K = k ] for each n , k and list the result in a table: n \ k 1 2 3 1 1 3 1 6 1 9 2 1 6 1 9 3 1 9 (c) Find the marginal distribution of N . Answer: The marginal distribution of N is given by the value of Pr [ N = n ] , for each possible value of n . By the total probability rule, Pr [ N = n ] = Pr [ N = n ∩ K = 1 ]+ Pr [ N = n ∩ K = 2 ]+ Pr [ N = n ∩ K = 3 ] . Thus, we get Pr [ N = n ] = 1 3 + 1 6 + 1 9 if n = 1 1 6 + 1 9 if n = 2 1 9 if n = 3 = 11 18 if n = 1 5 18 if n = 2 2 18 if n = 3 CS 70, Fall 2010, Soln 13 1 (d) Find the conditional distribution of K given that N = 1. Answer: By definition, Pr [ K = k  N = 1 ] = Pr [ K = k ∩ N = 1 ] Pr [ N = 1 ] . The numerator comes from the joint distribu tion of N and K (part (b)), and the denominator comes from the marginal distribution of N (part (c)). Plugging in, we get Pr [ K = k  N = 1 ] = 1 3 11 18 if k = 1 1 6 11 18 if k = 2 1 9 11 18 if k = 3 = 6 11 if k = 1 3 11 if k = 2 2 11 if k = 3 (e) We are now told that he bought at least 1 but no more than 2 books. Find the conditional mean and variance of K , given this piece of information. Answer: We first compute the distribution Pr [ K = k  N = 1 ∪ N = 2 ] as we did in part (d): Pr [ K = k  N = 1 ∪ N = 2 ] = 1 3 16 18 if k = 1 1 6 + 1 6 16 18 if k = 2 1 9 + 1 9 16 18 if k = 3 = 3 8 if k = 1 3 8 if k = 2 2 8 if k = 3 Now, the mean will be E ( K  N = 1 ∪ N = 2 ) = 1 × 3 8 + 2 × 3 8 + 3 × 2 8 = 15 8 and the variance will be Var ( K  N = 1 ∪ N = 2 ) = E ( K 2  N = 1 ∪ N = 2 ) E ( K  N = 1 ∪ N = 2 ) 2 = 1 2 × 3 8 + 2 2 × 3 8 + 3 2 × 2 8 15 8 2 = 39 64 ≈ . 61 (f) The cost of each book is a random variable with mean 3. What is the expectation of his total expendi ture? Hint: Condition on events N = 1 ,..., N = 3 and use the total expectation theorem....
View
Full
Document
This note was uploaded on 12/08/2010 for the course CS 70 taught by Professor Papadimitrou during the Fall '08 term at Berkeley.
 Fall '08
 PAPADIMITROU

Click to edit the document details