Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
MS125: Probability Assignment 2 To be submitted by 25th of April 2008 Question 1 The random variable x has the following discrete probability distribution: x P (X = x) 1 0.1 3 0.2 5 0.4 7 0.2 9 0.1 1. Write down cumulative probability function. 2. Calculate E[X]. 3. Calculate V ar[X]. Solution Cumulative probability function: F (x) = 0, 0.1, 0.3, 0.7, 0.9, 1, x < 1; 1 ≤ x < 3; 3 ≤ x < 5; 5 ≤ x < 7; 7 ≤ x < 9; x ≥ 9. E[X] = 1 ∗ 0.1 + 3 ∗ 0.2 + 5 ∗ 0.4 + 7 ∗ 0.2 + 9 ∗ 0.1 = 5 E[X 2 ] = 12 ∗ 0.1 + 32 ∗ 0.2 + 52 ∗ 0.4 + 72 ∗ 0.2 + 92 ∗ 0.1 = 29.8 V ar[X] = E[X 2 ] − E[X]2 = 29.8 − 25 = 4.8 Question 2 A machine that produces stampings for automobile engines is malfunctioning and producing 10% defectives. The defective and nondefective stampings proceed from the machine in a random manner. If the next five stampings are tested, find the probability that three of them are defective. 1 Solution Let X be number of defective machines. X ∼ B(5, 0.1) P (X = 3) = 53 0.13 0.92 = 10 ∗ 0.13 ∗ 0.92 = 0.0081 Question 3 Suppose the number, x, of a company’s employees who are absent on Mondays has a Poisson probability distribution. Furthermore, assume that the average number of Monday absentees is 2.6. 1. Find the mean and standard deviation of x, the number of employees absent on Monday. 2. Find the probability that fewer than two employees are absent on a given Monday. 3. Find the probability that more than three employees are absent on a given Monday. 4. Find the probability that exactly four employees are absent on a given Monday. Solution x ∼ P (2.6) 1. E[x] = 2.6 V ar[X] = 2.6, p √ Standard deviation: V ar[X] = 2.6 = 1.6125 2. P (x < 2) = P (x ≤ 1) = P (x = 0) + P (x = 1) = 0.0743 + 0.1931 = 0.2674 2.60 e−2.6 0! + 2.60 e−2.6 0! = 3. P (x > 3) = 1 − P (x ≤ 3) = 1 − (P (x = 0) + P (x = 1) + P (x = 2) + P (x = 1 −2.6 2 −2.6 3 −2.6 0 −2.6 e e e e + 2.6 1! + 2.6 2! + 2.6 3! ) = 1−(0.0743+0.1931+ 3)) = 1−( 2.6 0! 0.2510 + 0.2176) = 0.264 4. P (x = 4) = 2.64 e−2.6 4! = 0.1414. Question 4 Suppose the research department of a steel manufacturer believes that one of the company’s rolling machines is producing sheets of steel of varying thickness. The thickness is a discrete uniform random variable with values between 150 and 200 millimeters. Any sheets less than 160 millimeters must be scrapped because they are unacceptable to buyers. Calculate the fraction of steel sheets produced by this machine that have to be scrapped. 2 Solution Let X be the thickness of sheet. P P159 159 P (X < 160) = i=150 P (X = i) = i=150 1 200−150 = 1 50 P159 i=150 1= 10 50 = 0.2 Question 5 Assume the length of time, X, between charges of a mobile phone is normally distributed with a mean of 10 hours and a standard deviation of 1.5 hours. Find the probability that the cell phone will last between 8 and 12 hours between charges. Solution X ∼ N (10, 1.52 ) Z = X−10 1.5 ∼ N (0, 1) 12−10 P (8 < X < 12) = P ( 8−10 1.5 < Z < 1/5 ) = P (−1.33 < Z < 1.33) = P (Z < 1.33) − P (Z < −1.33) = 2 ∗ P (Z < 1.33) − 1 = 0.8164 Question 6 A fair dice is thrown three times. The result of first throw is scored as X1 = 1 if the dice shows 6 and X1 = 0 otherwise; X2 and X3 are scored likewise for the second and third throws. Let Y1 = X1 + X2 and Y2 = X1 − X3 . 1. Calculate the joint probabilities in the bivariate distribution of the pair (Y1 , Y2 ) and display in an appropriate table. 2. Find the marginal probability distributions of Y1 and Y2 . 3. Calculate the means and variances of Y1 and Y2 . 4. Calculate the covariance of Y1 and Y2 . 5. Find the conditional distribution of Y1 given Y2 = 0. 6. Find the conditional mean of Y1 given Y2 = 0. 7. State with justification if Y1 and Y2 are independent. Solution 1. P (X1 = 1) = P (X2 = 1) = P (X3 = 1) = P (X1 = 0) = P (X2 = 0) = P (X3 = 0) = 3 1 6 5 6 X1 0 0 0 0 1 1 1 1 Y2 X2 0 0 1 1 0 0 1 1 X3 0 1 0 1 0 1 0 1 Y1 = X1 + X2 0 0 1 1 1 1 2 2 0 Y1 1 2 P -1 25 216 5 216 0 30 216 0 125 216 30 216 1 216 156 216 1 0 25 216 5 216 30 216 P 150 216 60 216 6 216 1 Y2 = X1 − X3 0 -1 0 -1 1 0 1 0 5 6 5 6 5 6 5 6 1 6 1 6 1 6 1 6 × × × × × × × × 5 6 5 6 1 6 1 6 5 6 5 6 1 6 1 6 P × 56 × 16 × 56 × 16 × 56 × 16 × 56 × 16 = = = = = = = = 125 216 25 216 25 216 5 216 25 216 5 216 5 216 1 216 2. Marginal probability distribution of Y1 : y1 0 1 2 60 6 P (Y1 = y1 ) 150 216 216 216 Marginal probability distribution of Y2 : y2 -1 0 1 P (Y2 = y2 ) 30 216 156 216 30 216 3. E[Y1 ] = E[Y12 ] = V ar[Y1 ] = E[Y2 ] = E[Y22 ] = V ar[Y2 ] = 150 60 6 1 +1× +2× = 216 216 216 3 60 6 7 150 + 12 × + 22 × = 02 × 216 216 216 18 1 5 7 − = E[Y12 ] − (E[Y1 ])2 = 18 9 18 30 156 30 (−1) × +0× +1× =0 216 216 216 156 30 5 30 +0× + 12 × = (−1)2 × 216 216 216 18 5 E[Y22 ] − (E[Y2 ])2 = 18 0× 25 5 + 1 × 216 +2× 4. Cov[Y1 , Y2 ] = E[Y1 Y2 ] − E[Y1 ]E[Y2 ] = E[Y1 Y2 ] = 1(−1) 216 5 5 1 × 216 = 36 4 5. P (Y1 = 0|Y2 = 0) = P (Y1 = 1|Y2 = 0) = P (Y1 = 2|Y2 = 0) = 125/216 125 P (Y1 = 0, Y2 = 0) = = P (Y2 = 0) 156/216 156 P (Y1 = 1, Y2 = 0) 30/216 30 5 = = = P (Y2 = 0) 156/216 156 26 P (Y1 = 2, Y2 = 0) 1/216 1 = = P (Y2 = 0) 156/216 156 6. E[Y1 |Y2 = 0] = 0 + 1 × 1 8 5 +2× = 26 156 39 7. Y1 and Y2 are not independent because, for example, P (Y1 = 0, Y2 = 1) = 0 6= P (Y1 = 0)P (Y2 = 1) Question 7 x1 , . . . , xn is a random sample of size n from random variable with probability density function: f (x) = (1 + θ)xθ , 0<x<1 Find θ̂, the maximum likelihood estimator of θ. Solution Qn Qn Qn Qn θ θ M L = L(θ) = i=1 (1 + θ)xθi = ( Qi=1 (1 + θ)) (1 + θ)n ( i=1 xi ) i=1 xi = P n n l(θ) = ln L(θ) = n ln(1 + θ) + θ ln i=1 xi = n ln(1 + θ) + θ i=1 ln xi Pn dl(θ) n + i=1 ln xi = 0 dθ = 1+θ P n n i=1 ln xi 1+θ = − θ̂ = − Pn n ln x − 1 = −1/ln¯x − 1 i=1 i Question 8 A manufacturer of cereal wants to test the performance of one of its filling machines. The machine is designed to discharge a mean amount of 12 ounces per box, and the manufacturer wants to detect any departure from this setting. (i) He randomly samples 100 boxes from production run and obtains that the sample mean fill of 100 boxes is x̄ = 11.5 and standard deviation is s2 = 6.25. Test the null hypothesis H0 : µ = 12 against the alternative hypothesis H0 : µ 6= 12. (ii) The next day the manufacture samples 10 boxes and gets x̄ = 11.6 and s2 = 1.6. Test the null hypothesis H0 : µ = 12 against the alternative hypothesis H0 : µ 6= 12. 5 Solution (i) Single sample with unknown variance and large sample size. H0 : µ = 12 H0 : µ 6= 12 Two-tailed test. Test statistics: s2 =√6.25 s = 6.25 = 2.5 x̄−µ √ = 11.5−12 √ = −2 Z = s/ n 2.5/ 100 p = P (|Z| > 2) = 2 ∗ P (Z > 2) = 2 ∗ (1 − P (Z < 2)) = 2 ∗ (1 − 0.9772) = 0.0456 Moderate evidence against the null hypothesis that mean is equal to 12. (ii) Single sample with unknown variance and small sample size. H0 : µ = 12 H1 : µ 6= 12 s2 =√0.16 s = 0.16 = 0.4 Test statistics: x̄−12 √ = 11.6−12 √ = −3.16 T = s/ n 0.4/ 10 p = P (|T9 | > 3.16) = 2∗P (T9 > 3.16) = 2∗(1−P (T9 < 3.16)) = 2∗(1−0.9946) = 0.0108 There is a moderate evidence against the null hypothesis. Question 9 An economist decided to test the hypothesis that higher retail prices are being charged for Japanese automobiles in Japan than in the United States. She obtained random samples of 50 retail sales in the United States and 30 retail sales in Japan over the same time period and for the same model of automobile, converted the Japanese sales prices from yen to dollars using current conversion rates, and obtained the summary information shown in Table below. Sample size Sample mean Sample standard deviation USA Sales 50 $ 16,545 $ 1,989 Japan Sales 30 $ 17,243 $ 1,843 Test the null hypothesis H0 : µU = µJ against the alternative H0 : µU > µJ . Solution Two samples with unknown variances and large sample sizes. H0 : µU = µJ H1 : µU > µJ One-tailed test. Test statistics: = √ 16545−17243 Z = p 2x̄U −x̄J 2 = −1.59 2 2 Su /nU +Sj /nJ 1989 /50+1843 /30 p = P (|Z| > 1.59) = 1 − P (Z < 1.59) = 1 − 0.9441 = 0.0559 6 There is a weak evidence against the null hypothesis, mean retail prices are equal. Question 10 In order to create a behavioral profile of travelers, agency interviewed 5,026 travelers in the Tampa Bay region. Two of the characteristics they investigated were the travelers’ education level and their use of the Internet to seek travel information. The table below summarizes the results of the interviews. College Degree or More Less than a College Degree USE INTERNET Yes No 1,072 1,287 640 2,027 Test if travelers who use the Internet to search for travel information are likely to be people who are college educated. Solution H0 : The Internet use and the education are independent. H1 : The Internet use and the education are not independent. Observed frequencies: USE INTERNET Yes No 1072 1287 640 2027 1712 3314 College Degree or More Less than a College Degree Total Total 2359 2667 5026 Expected frequencies: College Degree or More Less than a College Degree USE INTERNET Yes No 803.54 1555.46 908.47 1758.54 Test statistics: 2 2 2 2 2 P4 i) χ2 = i=1 (Oi −E = (1072−803.54) + (1287−1555.46) + (640−908.47) + (2027−1758.54) = Ei 803.54 1555.46 908.47 1758.54 256.35 ν = (2 − 1)(2 − 1) = 1 p = P (χ21 > 256.35) P (χ21 > 12.12) = 0.0005 p < 0.0005 There a strong evidence against null hypothesis. The Internet use and education are not independent. 7