Sei sulla pagina 1di 5

Al akhawayne University School of Business Administration Advance Quantitative Methods GBU4401

Assignment: 1

Ahd Rankou Youssef El Hadaj Salim Nadifi

Dr. J.Abrache Spring 2007

I.

Problem 1 :

1. The probability that the sample mean is less than 0.75 seconds: P(X<0.75s)=? We have: Mean time =0.8 Standard deviation sigma = 0.2 Random sample: n= 30 We assume that the population is normally distributed: Z= (X u) / ( Sigma / n ) = (0.75 0.8) / (0.2 / 30) = 1.3698 From the table P (Z<1.36) = 0.0869 2. The probability that the sample mean is between 0.7 and 0.9 seconds: P (0.7<X <0.9) =? We compute the Z score: Z (0.7) = (0.7-0.8)/( 0.2 / 30) = -2.73 Z (0.9) = (0.9-0.8)/( 0.2 / 30) = +2.73 From the table: P(Z0.9 <2.73) = 0.9968 P(Z0.7 <-2.73) = 0.0032 P(0.7<X<0.9) = P(Z0.9 <2.73) - P(Z0.7 <-2.73) = 0.9936 3. The value of : is the distance between the mean and Xu ( upper tail) the probability say that we have 80% comprised between Xl and Xu which means that 10%is outside the interval divided into two equal parts of 5%. We compute the distance between the mean and Xu Xu = + (Z * (sigma / n) Xu = 0.8 +( 1.28 * (0.2/ 30) { Z(1.28) correspond to the probability of 10% in each tail) Xu = 0.8 +( 0.0467) So = 0.0467 4. If we reduce the sample size to 16, tha standard deviation of the sample mean will change as follow: We will have Sigma(x) = (0.2 / 16) = 0.05 instead of Sigma(x) = (0.2 / 30) = 0.0365 and we have to keep the assumption that the population is normally distributed.

II : Problem 2 : 1. The probability that the sample will be between 50% and 60% / We have: n = 200 = 0.5 P(0.50 < p < 0.60) = ? We check the assumptions: n = 100 >5 and n (1 -) = 100 >5 We compute the standard deviation: (p (1-p)/n) = (0.5 (1-0.5) / 200) = 0.0353 (P<0.6): Z = (X )/ (Sigma) = (0.6 0.5)/ (0.0353) = 2.832 (P<0.6) = (Z2.83 ) = 0.9977 (P<0.5) : Z = (X )/(Sigma)=(0.5 0.5 )/(0.0353 ) = 0 (P<0.5) = (Z0 ) = 0.5 P(0.50 < p < 0.60) = (Z2.83 ) - (Z0 ) = 0.4977 Thus the probability is : 0.4977 2. The probability that the sample proportion of correct identification is greater than 65%: (p>0.65)=? (P<0.65) : Z = (X )/(Sigma)=(0.65 0.5 )/(0.0353 ) = 4.24 (P<0.65) = (Z4.24 ) = We could not find it in the table. 3. Question 3: (P<0.6): Z = (X )/ (Sigma) = (0.6 0.5)/ ( (0.6(1-0.6)) 200) = 2.88 (P<0.55): Z = (X )/ (Sigma) = (0.55 -0.5)/ ( (0.55(1-0.45)) 1000) = 3.17 (P<0.6) = (Z2.88 ) = 0.9980 (P<0.55) = (Z3.17 ) =0.9992 60% from a sample of 200 is more likely to happen than 55% from a sample of 1000. The result is logical because if we increase the sample size we have more certitude to be close to the population.

III: Problem 3: 1. Construct a 99% confidence interval of the population mean: We have n= 70 X=1759 S=380 x=42 P= x/n = 42 /70 = 0.6 P=0.6 We use the t-distribution: X +/- t n-1/ 0.005 S / n = 1759 +/ - 2.6490*(380/70) We obtain: 1638.69 < Mean < 1879.31 We are 99% confident that the population mean is between 1638.69 and 1879.31 2. Construct a 99% confidence interval of the population proportion: We check the assumption n >5 and n (1- )>5 Since we dont have we can use P to approximate. nP>5 and n(1-P)>5 We compute the standard deviation for the proportion: Sigma = (p (1-p)/n) = (0.6(1-0.6)/70) P +/- Z * (p (1-p)/n) 0.6 +/- 1.96 * (0.6(1-0.6)/70) 0.48 < < 0.71 We are 95% confident that the population proportion of houses that have the air conditioning is between 0.48 and 0.71 IV Problem 4: 1. Set a 90 % confidence interval: We check the assumption n >5 and n (1- )>5 Since we dont have we can use P to approximate. nP >5 and n(1-P)>5 We have: n = 300; P +/- Z (p (1-p)/n) (30/ 300)+/- 1.64 * (0.1(1-0.1)/300) {1.64 is the Z value corresponding to 0.05) 0.072 < < 0.128 We are 90% confident that the estimate proportion of deficient pens in the shipment is between 7.2% and 12.8%.

2. Question 2: According to question one, we are 90% confident that the number of deficient pens is between 7.2% and 12.8% which more than 5%. The manager can return the shipment. 3. Set a 99% confidence interval: P +/- Z (p (1-p)/n) (30/ 300)+/- 2.57 * (0.1(1-0.1)/300) {2.57 is the Z value corresponding to 0.005) 0.055 < < 0.14 We are 99% confident that the estimate proportion of deficient pens in the shipment is between 5.5% and 14%. We keep the decision that the manager should return the shipment because the number of deficient pens is still more than 5%. V: Problem 5 1- The sample size needed if the agency wants to be 90% confident: n = [(Z/2) * ]/ error n = 220 2- The necessary sample size to be 99 % confident: n = [(Z/2) * ]/ error n = 538 VI: Problem 6 = $18 1-The sample size needed to have 99% confidence of estimating the population mean : n = [(Z/2) * ]/ error = (2.575 * 18)/5 = 85.9329 n = 86 2- The necessary sample size to be 90 % confident: n = [z * *(1 )] / error = (1.645 * 0.05 * 0.5)/0.045 = 334.07 n = 335 3. We will sample the number of stores customers that that we need for each purpose. = (2.575 * 45)/5 = 537.0806 = (1.645 * 45)/5 = 219.188

Potrebbero piacerti anche