STAT 571 Qi Jiang [email protected] Solution for Homework 6 Problem 1 In this question, we use two R commands: pnorm() and qnorm() to caculate the cumulative probability and quantile of normal distributed random variable respectively. 1. use R command pnorm() > pnorm(0.75) [1] 0.7733726 Using the normal table in the textbook on Page 673, we know P (Z ≥ 0.75) = 0.2663, so P (Z ≤ 0.75) = 1 − P (Z ≥ 0.75) = 0.7337 2. use R command pnorm() > 1-pnorm(-0.97) [1] 0.8339768 Since standard normal distribution is symmetric P (Z ≥ −0.97) = P (Z ≤ 0.97) = 1 − P (Z ≥ 0.97). Using the normal table in the textbook on Page 673, we have P (Z ≥ 0.97) = 0.16602, then P (Z ≥ −0.97) = 1 − P (Z ≥ 0.97) = 0.83398 3. use R command pnorm() > pnorm(-1.11)-pnorm(-1.18) [1] 0.01449941 P (−1.18 ≤ Z ≤ −1.11) = P (1.11 ≤ Z ≤ 1.18) = P (Z ≤ 1.18) − P (Z ≤ 1.11) = (1 − P (Z ≥ 1.18)) − (1 − P (Z ≥ 1.18)) Using the normal table in the textbook on Page 673, we have P (Z ≥ 1.18) = 0.11900 and P (Z ≥ 1.11) = 0.13350. So, P (−1.18 ≤ Z ≤ −1.11) = (1 − 0.11900) − (1 − 0.1335) = 0.0145 4. Since P ()Z > z) = 0.05 implies that P (Z ≤ z) = 0.95, use R command qnorm() > qnorm(0.95,0,1) [1] 1.644854 So z = 1.644854 Using the normal table in the textbook on Page 673, we find the correspondent value that makes the right tail probability almost equals to 0.05. z is between 1.64 and 1.65. 5. use R command qnorm() > qnorm(0.1,0,1) [1] -1.281552 P (Z < z) = 0.10 implies that z is a negative number. However we can only look up positive numbers in the normal distribution table. Note that P (Z > −z) = 0.1, so we can find −z instead. Using the normal table in the textbook on Page 673, we find −z is 1.28. Thus z is −1.28. Office: 1270 MSC 1 STAT 571 Qi Jiang [email protected] 6. P (−z < Z < z) = 0.7 implies that P (Z ≤ z) = 0.85. use R command qnorm() > qnorm(0.85,0,1) [1] 1.036433 P (Z ≤ z) = 0.85, so that P (Z > z) = 0.15 Using the normal table in the textbook on Page 673, we find that z = 1.04. Problem 2 In the following problems, we use two related R commands:pt() and qt(). 1. Use R command > pt(0.75,10) [1] 0.764734 In the t distribution table, we know that the 10% right tail quantile is 1.37, which is larger than 0.75. Thus we can make a rough estimation that P (T ≤ 0.75) is less than 0.9. 2. P (T ≥ −0.97) = 1 − P (T < −0.97) Use R command > 1-pt(-0.97,13) [1] 0.8251257 Since P (T ≥ −0.97) = P (T ≤ 0.97) and the 10% quantile is 1.35, which is larger than 0.97, so the rough estimation is less than 0.9. 3. P (−1.18 ≤ T ≤ −1.11) = P (T ≤ −1.11) − P (T ≤ −1.18) Use R command > pt(-1.11,19)-pt(-1.18,19) [1] 0.01414329 P (−1.18 ≤ T ≤ −1.11) = P (1.11 ≤ T ≤ 1.18). In the t distribution table, both 1.11 and 1.18 is between 0(50% quantile) and 10% quantile, so P (−1.18 ≤ T ≤ −1.11) is between 0.5 and 0.9. 4. P (T > z) = 0.05 implies that P (T ≤ z) = 0.95 Use R command > qt(0.95,11) [1] 1.795885 So z = 1.795885. In this case, we find the one tail 5% critical value for a t distribution with df of 11 is 1.80. 5. Use R command > qt(0.1,15) [1] -1.340606 So z = −1.340606. Since P (Z < z) = 0.1 < 0.5 implies z is negative, we would find −z instead. Note P (Z > −z) = 0.1. In the t table, we find the one tail 10% critical value for a t distribution with a df of 15 is 1.34. Thus z = −1.34. Office: 1270 MSC 2 STAT 571 Qi Jiang [email protected] 6. P (−z < Z < z) = 0.7 implies that P (Z ≤ z) = 0.85. Use R command > qt(0.85,17) [1] 1.069033 So z = 1.069033. Also P (−z < Z < z) = 0.7 implies that P (Z > z) = 0.15, but there is no 15% critical value for t distribution, so we cannot find it in the t table. Problem 3 We know that X follows a normal distribution with parameters µ = 1.25 and σ = 0.0080. The sample size n = 25. 1. we want find P (1.26 ≤ X ≤ 1.30 = P (X ≤ 1.30) − P (X < 1.26). Use R, > pnorm(1.30,1.25,0.0080)-pnorm(1.26,1.25,0.0080) [1] 0.1056498 Thus P (1.26 ≤ X ≤ 1.30 = 0.1056498. 2. we want find z such that P (X ≥ z) = 0.65. Note P (X ≥ z) = 0.65 implies that P (X < z) = 0.35. Use R, > qnorm(0.35,1.25,0.008) [1] 1.246917 Thus z = 1.246917. 3. Since X is normal distribution N (µ, σ), thus X̄ is N (µ, √σn ) which is N (1.25, 0.0016). Use R, >pnorm(1.26,1.25,0.0016)-pnorm(1.24,1.25,0.0016) [1] 1 4. Use R, > qnorm(0.9,1.25,0.0016) [1] 1.25205 So the quantile is 1.25205. 5. Let the cut off values be −z and z. Thus P (X̄ ≤ z) = 0.9. Use R, > qnorm(0.9,1.25,0.0016) [1] 1.25205 Thus the cut off values are −1.25205 and −1.25205. Office: 1270 MSC 3 STAT 571 Qi Jiang [email protected] Problem 4 Let X denote the traffic fatalities for each U.S. state resulting from drivers with high alcohol blood levels in 1982. Then X is normal distributed with parameters of µ = 0.569 and σ = 0.068. 1. we want to find P (X > 0.65). Use R, > 1-pnorm(0.65,0.569,0.068) [1] 0.1167922 So P (X > 0.65) = 0.1167922. 2. > qnorm(0.25,0.569,0.068) [1] 0.5231347 So the 25th quantile of the fatalities is 0.5231347. Problem 5 To calculate P (Y > y), we use commands 1 − pnorm(y, µ, σ), and to calculate P (Y < y), we just use pnorm(y, µ, σ). 1. For µ = 14 and σ = 5, y = 9, > 1-pnorm(9,14,5) [1] 0.8413447 > pnorm(9,14,5) [1] 0.1586553 P (Y > y) = 0.8413447, P (Y < y) = 0.1586553 2. For µ = 15 and σ = 3, y = 18.5, > 1-pnorm(18.5,15,3) [1] 0.1216725 > pnorm(18.5,15,3) [1] 0.8783275 P (Y > y) = 0.1216725, P (Y < y) = 0.8783275 3. For µ = −23 and σ = 4, y = −16, > 1-pnorm(-16,-23,4) [1] 0.04005916 > pnorm(-16,-23,4) [1] 0.9599408 P (Y > y) = 0.04005916, P (Y < y) = 0.9599408. 4. For µ = 14000 and σ = 5000, y = 9000, Office: 1270 MSC 4 STAT 571 Qi Jiang [email protected] > 1-pnorm(9000,14000,5000) [1] 0.8413447 > pnorm(9000,14000,5000) [1] 0.1586553 P (Y > y) = 0.8413447, P (Y < y) = 0.1586553. Problem 6 X̄ is distributed as N (µ, √σn ) To calculate P (Y > y), we use commands 1 − pnorm(y, µ, σ), and to calculate P (Y < y), we just use pnorm(y, µ, σ). 1. For µ = −5 and σ = 5, y = −5.2, > pnorm(-5.2,-5,5/sqrt(20)) [1] 0.4290138 > pnorm(-5.2,-5,5/sqrt(50)) [1] 0.3886487 #For n=20 #For n=50 Thus P (Ȳ < y) = 0.4290138 when n = 20; P (Ȳ < y) = 0.3886487 when n = 50; 2. For µ = 10 and σ = 30, y = 8, > pnorm(8,10,30/sqrt(20)) [1] 0.3827972 > pnorm(8,10,30/sqrt(50)) [1] 0.3186759 #For n=20 #For n=50 Thus P (Ȳ < y) = 0.3827972 when n = 20; P (Ȳ < y) = 0.3186759 when n = 50; 3. For µ = −55 and σ = 20, y = −61, > pnorm(-61,-55,20/sqrt(20)) [1] 0.08985625 > pnorm(-61,-55,20/sqrt(50)) [1] 0.01694743 #For n=20 #For n=50 Thus P (Ȳ < y) = 0.08985625 when n = 20; P (Ȳ < y) = 0.01694743 when n = 50; 4. For µ = 12 and σ = 3, y = 12.5, > pnorm(12.5,12,3/sqrt(20)) [1] 0.7719717 > pnorm(12.5,12,3/sqrt(50)) [1] 0.8807036 #For n=20 #For n=50 Thus P (Ȳ < y) = 0.7719717 when n = 20; P (Ȳ < y) = 0.8807036 when n = 50; Office: 1270 MSC 5
© Copyright 2026 Paperzz