Vous êtes sur la page 1sur 5

# NAME: MINH NGUYEN Class: MATH 533 Project Part B In this report, I will be performing hypothesis testing to each

requirement using seven elements of Test Hypothesis with a = 0.02 in order to provide more insights for my manager. I will also include the interpretation of results so that non-statistical people can understand the meaning of number. a) The average (mean) annual income of all customers was less than \$48,000

I found the average annual income is \$43.74 and the standard deviation is \$14.64. Step1: Null and Alternative Hypothesis Set up Null Hypothesis: The average (mean) annual income was greater than or equal to \$48000 Ho: >= 48 Alternate Hypothesis: The average (mean) annual income was less than \$48000. Ha: < 48 Step2: For =0.02. Since the sample size, n > 30 we will use z-test for mean to test the given hypothesis. As the alternative hypothesis is Ha: < 48, the given test is a one-tailed (lower-tailed) ztest. Step3: The critical value for significance level, =0.02, I found that z = 1-.960. Therefore, the rejection region will be z < -1.960 Step4: Test Statistic (MINITAB OUTPUT):
Test of mu = 48 vs < 48 The assumed standard deviation = 14.64 98% Upper Bound 47.99

## Variable Income (\$1000)

N 50

Mean 43.74

StDev 14.64

SE Mean 2.07

Z -2.06

P 0.020

Step5: Interpretation Since the P-value 0.02 is less the significance level 0.025. In addition, z value falls into the rejection region which is -1.960. Therefore, we reject the null hypothesis. At a significance level of 0.02, there is sufficient evidence to support the claim that the average (mean) annual income was less than \$48000. Confidence Interval (MINITAB OUTPUT):
The assumed standard deviation = 14.64 Variable N Mean StDev SE Mean 98% CI

Income (\$1000)

50

43.74

14.64

2.07

(38.92, 48.56)

According to the confidence level, we are 98% sure that the mean income lies between \$38,920 and \$48,560.

b)

The average (mean) annual income of urban customers was more than \$40,000

I found the average annual income is \$44.95 and the standard deviation is \$14.56. Step1: Null and Alternative Hypothesis Set up Null Hypothesis: The average (mean) annual income was smaller than or equal to \$40000 Ho: <= 40 Alternate Hypothesis: The average (mean) annual income was more than \$40000. Ha: > 40 Step2: For =0.02. We will use z-test for mean to test the given hypothesis. As the alternative hypothesis is Ha: > 40, the given test is a one-tailed (upper-tailed) z-test. Step3: The critical value for significance level, =0.02, I found that z = 1.960. Therefore, the rejection region will be z > 1.960 Step4: Test Statistic (MINITAB OUTPUT):
Test of mu = 40 vs > 40 The assumed standard deviation = 14.56 98% Lower Bound 38.57

N 22

Mean 44.95

SE Mean 3.10

Z 1.59

P 0.055

Step5: Interpretation Since the P-value 0.05 is greater than 0.025. In addition, z value doesnt fall into the rejection region which is 1.960. Therefore, we fail to reject the null hypothesis. At a significance level of 0.02, there is not sufficient evidence to support the claim that the average (mean) annual income of urban customers was more than \$40000. Confidence Interval (MINITAB OUTPUT):
Test of mu = 40 vs not = 40 The assumed standard deviation = 14.56 N 22 Mean 44.95 SE Mean 3.10 98% CI (37.73, 52.17) Z 1.59 P 0.111

According to the confidence level, we are 98% sure that the mean income of urban customers lies between \$37,730 and \$52,170.

c. The true population proportion of customers who live in an urban area exceeds

42%
Step1: Null and Alternate Hypothesis Set up: Null Hypothesis: The true population proportion of customers who live in an urban area is equal to 42%. Ho: p = 0.42 Alternate Hypothesis: The true population proportion of customers who live in an urban area is greater than 42%. Ha: p > 0.42 Step2: =0.02. There are 22 people live in the Urban area, which is 44% or 0.44. This is the point estimate for p. As the alternative hypothesis is Ha: p > 0.42. The critical value for significance level, =0.02, we first need to verify that the sample size is large enough. nPo = 50 * 0.42 = 21 and 50 * (1 0.42) = 29. Both are larger than 15 so we can conclude that the sample size is large enough to apply the large sample z test. Step3: Rejection region for one-tailed Z test will be z > 1.960 Step4: Test Statistic (MINITAB OUTPUT): Test and CI for One Proportion
Test of p = 0.42 vs p > 0.42 98% Lower Bound 0.295827

Sample 1

X 22

N 50

Sample p 0.440000

Z-Value 0.29

P-Value 0.387

## Using the normal approximation.

Step5: Interpretation Since the P-value is 0.387, which is greater than the significance level 0.025, we fail to reject the null hypothesis. There is no sufficient evidence to support the claim that the true population proportion of customers who live in an urban area is greater than 42%. Test and CI for One Proportion
Test of p = 0.42 vs p not = 0.42 Sample 1 X 22 N 50 Sample p 0.440000 98% CI (0.276691, 0.603309) Z-Value 0.29 P-Value 0.774

The 98% lower confidence limit is 0.27. Since 0.42 is greater than the 98% lower confidence limit, hence we cannot support the claim that the true population proportion of customers who live in an urban area is greater than 42%. We are 98% confident that the mean population lies between 0.276 and 0.603.

d. The average (mean) number of years lived in the current home is less than 10 years. Step1: Null and Alternate Hypothesis Set up: Null Hypothesis: The average (mean) number of years lived in the current home is equal 10 years Ho: = 10 Alternate Hypothesis: The average (mean) number of years lived in the current home is less than 10 years. Ha: < 10 Step2: Since the sample size we shall use z-test for mean to test the given hypothesis. As the

alternative hypothesis is Ha < 10 the given test is a one-tailed (lower-tailed) z-test. Step3: For =0.02, rejection region will be z < -1.960. Step4: Test Statistic (MINITAB OUTPUT):
Test of mu = 10 vs < 10 The assumed standard deviation = 5.086 98% Upper Bound 13.737

Variable Years

N 50

Mean 12.260

StDev 5.086

SE Mean 0.719

Z 3.14

P 0.999

Step5: Interpretation Since the P-value 0.999 is greater than the significance level 0.025, we fail to reject the null hypothesis. There is no sufficient evidence to support the claim that the average (mean) number of years lived in the current home is less than 10 years. Confidence Interval (MINITAB OUTPUT):
Variable Years N 50 Mean 12.260 StDev 5.086 SE Mean 0.719 98% CI (10.587, 13.933) Z 3.14 P 0.002

The 98% upper confidence limit is 13.933. Since, 10 is smaller than the 98% upper confidence limit, hence we cannot support the claim that the average (mean) number of years lived in the current home is less than 10 years. We are 98% confident that mean number of years lies between 10.587 and 13.933. e. The average (mean) credit balance for suburban customers is more than \$4000 Step1: Null and Alternate Hypothesis Set up: Null Hypothesis: The average (mean) credit balance for suburban customers is equal to \$4000 Ho: = 4000 Alternate Hypothesis: The average (mean) credit balance for suburban customers is more than \$4000. Ha: > 4000

## we shall use z-test for mean to test the given hypothesis.

As the alternative hypothesis is Ha > 4000 the given test is a one-tailed (upper-tailed) z-test. Step3: For =0.02, I found the rejection rule is z > 1.960 Step4: Test Statistic (MINITAB OUTPUT):
Test of mu = 4000 vs > 4000 The assumed standard deviation = 932 98% Lower Bound 3700

## Variable Credit Balance(\$)

N 50

Mean 3970

StDev 932

SE Mean 132

Z -0.22

P 0.589

Step5: Interpretation Since the P-value 0.589 is larger than the significance level 0.02, we fail reject the null hypothesis. Thus, at a significance level of 0.02, there is not sufficient evidence to support the claim that the average (mean) credit balance for suburban customers is more than \$4000. Confidence Interval (MINITAB OUTPUT):
Variable Credit Balance(\$) N 50 Mean 3970 StDev 932 SE Mean 132 98% CI (3664, 4277) Z -0.22 P 0.823

One-Sample Z The assumed standard deviation = 770 N 50 Mean 4654 SE Mean 95% CI 109 (4441, 4867)

The 98% lower confidence limit is 3664. Since 4000 is larger than the 98% lower confidence limit, hence we can support the claim that the average (mean) credit balance for suburban customers is not more than \$4000. We are 98% confident that the mean credit balance lies between \$3664 and \$4277.