Stat560 Solution to Final Exam 1. True/False. Reason the answers. (a) If two continuous random variables jointly have a rectangular support then they must be independent. Answer: False. Counterexample: for X and Y having joint p.m.f. as follows, the support is rectangular. However, they are dependent. y x 0 1 0 1 fX (x) 1 4 1 6 1 4 1 3 1 2 1 2 fY (y) 5 12 7 12 See also Problem 5 for another counterexample. (b) The probability that a continuous random variable assumes a particular value is zero. Answer: True. 2. (Multiple Choise). Reason the answer. Assume random variable X has mean 3 and variance 4. If Y = −X + 5, then a) µY = 2, σY2 = 1 b) µY = 8, σY2 = 4 c) µY = 2, σY2 = 4 d) µY = 8, σY2 = 1 2 Answer: (c). µY = −µX + 5 = 2. σY2 = σX = 4. 3. Suppose that the random variable X be exponentially distributed with parameter λ = 2. Use any method to determine the distribution of Y = exp(−λX). 1 distribution function method: For 0 < y < 1, FY (y) = P (Y ≤ y) = P (exp(−2X) ≤ y) = P (−2X ≤ lny) lny lny = P (X ≥ − ) = exp −2 · (− ) = y. 2 2 Consequently, fY (y) = FY0 (y) = 1 for 0 < y < 1. That is, Y ∼ U (0, 1). transformation method: For 0 < y < 1, y = exp(−2x) ⇒ x = − lny 2 1 dx =− and hence dy 2y lny 1 fY (y) = fX (− )− 2 2y 1 = 2y · = 1. 2y 4. According to a genetics theory, a certain cross of guinea pigs will result in red, black, and white offspring in the ratio 8:4:4. (a) Find the probability that among eight offspring five will be red, two black, and one white. (Hint: multinomial distribution) 8 The probabilities ofan offspring being red, black, and white are pr = 8+4+4 = 12 , 4 = 14 , respectively. Hence the probability that among eight offspring pb = pw = 8+4+4 five will be red, two black, and one white is 5 2 8 1 1 1 21 = 0.0820. · · = 5, 2, 1 2 4 4 256 (b) Of 50 offspring what is the probability that at least 30 will be either black or white? Use normal approximation. (Hint: binomial distribution) 2 Of 50 offspring, N , the number of black or white ones follows a binomial distribution b(n = 50, p = 12 ). The mean and the standard deviation of this number are, respecq 1 tively, µ = 50 · 2 = 25, and σ = 50 · 12 · 12 = 3.5355. Hence the desired probability can be approximated by P (N ≥ 30) = P (N > 29.5) continuity correction 29.5 − 25 ≈ P Z> = P (Z > 1.27) = 1 − Φ(1.27) 3.5355 = 1 − 0.8980 = 0.1020. 5. If the joint distribution of X and Y is given by y 1 2 3 x 0 a 1 2 1 8 1 32 1 32 a 1 4 3 16 3 a a 1 32 a 1 32 (a) Find the value of a in the table. a = 15 (1 − 81 − 14 − 3 36 −4· 1 ) 32 = 1 . 16 (b) Compute the marginal p.m.f. of X. The row margins in the table below show the p.m.f. of X: y 1 x 2 3 fY (y) 0 1 2 3 fX (x) 1 16 1 32 1 32 1 8 1 8 1 16 1 32 7 32 1 4 3 16 1 16 1 2 1 16 1 16 1 32 5 32 1 2 11 32 5 32 (c) Are X and Y independent? Although the support of (X, Y ) is rectangular, X and Y are dependent since, for instance, 1 1 7 f (1, 1) = 6= · = fX (1)fY (1). 8 2 32 3