1. The diameter of a metal cylinder is a random variable X with a probability density function given by fX (x) = c[1 − 4(x − 50)2 ], 49.5 ≤ x ≤ 50.5. (a) What is the value of c that makes fX (x) a probability density function? (5%) Ans. c = 32 . Sol) Z 1 = ·Z 50.5 2 c Z 50.5 [1 − 4(x − 50) ]dx = c dx − 4 49.5 " # ¯.5 · ¸ Z .5 4 ¯ 2 c 1−4 x2 dx = c 1 − x3 ¯¯ = c 3 3 −.5 −.5 49.5 = ⇒c = 50.5 ¸ (x − 50) dx 2 49.5 3 2 (b) What is the probability that the diameter X is less than or equal to 50, using your answer to part (a)? (5%) Ans. P (X ≤ 50) = 0.5. Sol) P (X ≤ 50) = 0.5 (since fX is symmetrical to X = 50.) 2. Suppose that X has a exponential distribution with p.d.f. f (x) = λe−λx , x > 0. Find the p.d.f. of Y = X 2 . (10%) Ans. fY (y) = λ2 y −1/2 e−λy 1/2 , y > 0. Sol) Fact: Y is monotonically increasing with respect to X. Let Y = g(X) = X 2 . Then, I(y) = (0, ∞), and ¯ ¯ ¡ −1 ¢ ¯ d −1 ¯ ¯ fY (y) = fX g (y) · ¯ g (y)¯¯ dy • g −1 (y) = y 1/2 . = 12 y −1/2 . ¯ ¯ ¯ d −1 ¯ 1 −1/2 • ¯ dy g (y)¯ = 2 y . ¡ ¢ 1/2 • fX g −1 (y) = λe−λy . • d −1 (y) dy g • So, fY (y) = λ −1/2 −λy1/2 y e , 2 y > 0. 3. Let Y have a continuous probability distribution with p.d.f. fY (y) = ye−y , 0 < y < ∞; then let X have a conditional distribution (given Y ) that is uniform on the interval from 0 to Y . 1 (a) Give the joint density function for X and Y ; be sure to give your answer correctly for all x and y. (5%) Ans. f (x, y) = fX|Y (x|y)fY (y) = e−y , 0 < x < y < ∞.. Sol) From the problem setting, we have fX|Y (x|y) = Hence, 1 , y 0<x<y<∞ f (x, y) = fX|Y (x|y)fY (y) = e−y , 0 < x < y < ∞. (b) Find the marginal density functions for X and Y . (5%) Ans. fX (x) = e−x , 0 < x < ∞ and fY (y) = ye−y , 0 < y < ∞. Sol) Z fX (x) ∞ = Zxy fY (y) = ¯∞ e−y dy = −e−y ¯x = e−x , e−y dx = ye−y , 0 < x < ∞. 0 < y < ∞. 0 (c) Find the conditional distribution of X given Y . (5%) Ans. fX|Y = ex−y , 0 < x < y < ∞. Sol) fX|Y = f (x, y) e−y = −x = ex−y , fX (x) e 0 < x < y < ∞. 4. The joint density function of X and Y is ½ f (x, y) = x+y 0 0 < x < 1, 0 < y < 1 otherwise Find P (X + Y < 1). (5%) Ans. P (X + Y < 1) = 13 . Sol) ¶y=1−x µ 1 dx xy + y 2 2 0 0 0 y=0 µ ¶x=1 Z 1 1 1 1 1 (1 − x2 )dx = x − x3 = 2 0 2 3 3 x=0 Z P (X + Y < 1) = = 1 Z Z 1−x 1 (x + y)dydx = 5. Suppose that the life distribution of an item has hazard rate function h(t) = t3 , t > 0. Find the probability that 2 (a) the item survives to age 2; (5%) Ans. e−4 = 0.01832. Sol) Denote the lifetime of the item as X. Then, Z t Z t 1 H(t) = h(x)dx = x3 dx = t4 4 0 0 R(t) ⇒ P (X > 2) 1 4 = e−H(t) = e− 4 t , t > 0 = R(2) = e−4 = 0.01832 (b) the item’s lifetime is between .4 and 1.4; (5%) 1 4 4 1 Ans. e− 4 ×0.4 − e− 4 ×1.4 = 0.6109. Sol) F (t) = ⇒ P (.4 < X < 1.4) = 1 4 1 − R(t) = 1 − e− 4 t , F (1.4) − F (0.4) = e t>0 − 14 ×0.44 1 4 − e− 4 ×1.4 = 0.9936 − 0.3827 = 0.6109 (c) the 1-year-old item will survive to age 2. (5%) Ans. e−15/4 = 0.02352. Sol) 1 P (X > 2|X > 1) = 4 P (X > 2 and X > 1) P (X > 2) e− 4 ×2 = = − 1 ×14 = e−15/4 = 0.02352 P (X > 1) P (X > 1) e 4 6. A randomly chosen IQ test taker obtains a score that is approximately a normal random variable with mean 100 and standard deviation 15. What is the probability that the test score of such a person is (a) above 125; (5%) Ans. 0.04779. Sol) Denote the score of the test as X. Since X ∼ (100, 152 ), we have µ ¶ 125 − 100 P (X > 125) = 1 − Φ = 1 − Φ(1.667) = 1 − 0.95221 = 0.04779 15 (b) between 90 and 110. (5%) Ans. 0.4952. Sol) ¶ µ ¶ 110 − 100 90 − 100 −Φ P (90 < X < 110) = Φ 15 15 = Φ(0.667) − Φ(−0.667) = 2Φ(0.667) − 1 µ = 2 × 0.7476 − 1 = 0.4952 7. If X1 and X2 are independent exponential random variables each having parameter λ, find the joint density function of Y1 = X1 + X2 and Y2 = eX1 . (10%) Ans. fY1 ,Y2 (y1 , y2 ) = λ2 −λy1 , y2 e 0 < ln y2 < y1 < ∞. Sol) The joint density function of X1 and X2 is fX1 ,X2 (x1 , x2 ) = λ2 e−λ(x1 +x2 ) , fY1 ,Y2 (y1 , y2 ) can be found through the following steps: 3 x1 , x2 > 0. • Expressing X1 and X2 in terms of Y1 and Y2 , then X1 X2 = = • The Jacobian matrix is then · J= 0 1 ln Y2 Y1 − ln Y2 1/y2 −1/y2 ¸ • From the definition of Y1 and Y2 , we have | det(J)| = 1/y2 • Then, fY1 ,Y2 (y1 , y2 ) = | det(J)| · fX1 ,X2 (ln y2 , y1 − ln y2 ) λ2 −λy1 = e , 0 < ln y2 < y1 < ∞ y2 8. Suppose that the number of fishes I catch up to time t is a Poisson process Xt with rate λ = 10 fishes/hour. (a) What is the probability that I catch no fish in the first 15 minutes? (5%) Ans. e−2.5 = 0.08209. Sol) Let X0.25 represent the number of fish being catched within the 0.25 hour (i.e., 15 minutes). Then, X0.25 ∼ P (2.5). Hence, 2.50 e−2.5 P (X0.25 = 0) = = e−2.5 = 0.08209. 0! (b) Given that I catch 3 fishes in the time period 6:00-6:15, what is the probability I’ll catch 5 fishes in the time interval 6:00-6:30? (5%) Ans. 3.125e−2.5 = 0.2565. Sol) The desired probability is the same as the probability to catch 2 fishes in time interval 6:15-6:30, i.e., in 15 minutes. Hence, it is P (X0.25 = 2) = 2.52 e−2.5 = 3.125e−2.5 = 0.2565. 2! (c) Let T10 be the time it takes for me to catch 10 fish; T10 is of course a random variable. How is T10 distributed? In particular, write down its cumulative distribution function F10 (t) expressed either as a sum or integral. (5%) Z t Ans. T10 ∼ Γ(10, 10), and F10 (t) = 0 9 X (10t)k e−10t 1010 x9 e−10x dx = 1 − , 9! k! t > 0. k=0 Sol) It is clear that T10 ∼ Γ(10, 10). So, f10 (t) = Hence, Z t F10 (t) = 0 1010 t9 e−10t , 9! t>0 1010 x9 e−10x dx, 9! t>0 By relating gamma distribution with Poisson distribution, we have Z F10 (t) = 0 t 9 X (10t)k e−10t 1010 x9 e−10x dx = 1 − , 9! k! k=0 9. Evaluate the following expressions: 4 t > 0. Z 10 (a) x4 e−x/2 dx =? (5%) 0 Z 10 Ans. 4 −x/2 x e 0 4! dx = 0.55 à 4 X 5k e−5 1− ! = 429.7012 k! k=0 Sol) Fact: λr (r − 1)! Z t xr−1 e−λx dx = 1− 0 Z k=0 t ⇒ r−1 −λx x e dx = x4 e−x/2 dx = 0 Z 10 ⇒ 0 Z ∞ (b) 2 e−(x−9) 9 Z ∞ Ans. /32 r−1 X (λt)k e−λt k! à ! r−1 X (r − 1)! (λt)k e−λt 1− λr k! k=0 ! à 4 X 5k e−5 4! = 429.7012 1 − 0.55 k! k=0 dx =? (5%) 2 e−(x−9) /32 dx = √ 8π. 9 Sol) Fact: 1 √ 2πσ Z ∞ µ ∞ e−(x−µ) 2 /(2σ 2 ) dx e−(x−µ) 2 /(2σ 2 ) dx 1 2 p π/2σ = dx = Z ⇒ µ Z ⇒ ∞ = Z 2 e−(x−9) 9 /32 ∞ 9 5 2 e−(x−9) /(2×42 ) dx = √ 8π P?@ 9 ½ ¾ ¾ ºººººººººººººººººººººººººº ºººººº º ººººººº ººººº º º ºººººº ººººº º º º ººººººº º º º ºººº º º º º ºººººº ºººº ºººº ºººº ºººº º º º º º ºººº ººººº º º º º º ºººº ºººº º º º º º º º º ºººº ººººº º º º º º º º º º º º º ºººº º º º ºº º º º º º º º ºººº º º º º ºººº ºººº º º º º º º º º ººººº º º ºººº º º º º º º º º ºººººº º º º º º ºººººº ºººº º º º º º º º º º º º ºººººººººº º º º ºººººººººººººººººº ººººººº º º º º º º º º º º º º º º º º º º º º º º º º º º º ººººººººººººººººººººººººººººººººººººººººººº ºººººººººººººººººººººººººººººººººººººººººº 6
© Copyright 2024 Paperzz