You are on page 1of 10

Vietnam National University – HCMC Semester 01, Academic year 2013 - 2014

INTERNATIONAL UNIVERSITY Course: Statistics for Business

---OOO---
FINAL-TERM EXAMINATION
Duration:
SOLUTION TO SAMPLE TEST 03 150 Minutes

Student ID: Date:

Name:
1. INSTRUCTIONS
2. This is an open book examination.
3. Calculator and dictionary are allowed.
4. Discussion and material transfer are strictly prohibited.
PART I : (Please answer on the QUESTION SHEET)

1. VinaGame makes a sample to see rate of student playing game online. The sample of 200
students was conducted, with sample proportion of 67%. Construct the confident interval for
with 90%. (3 points)
a) [0.4523 ; 0.6453] d) [0.6153 ; 0.7247]
b) [ 0.6048 ; 0.7352] e) None of the above, specify __________
c) [0.5153 ; 0.7532]

HINT:
̂ (1 − ̂ ) 0.67(1 − 0.67)
90% = ̂± / = 0.67 ± 1.645 ≈ [0.6153 ; 0.7247]
200

2. BigC CCO wants to estimate the population for customer waiting in line. A sample of 25
customer constructed with = 4 minutes. Give 95% confident interval for population
variance (assumed population is normal distributed) (3 points)
a) [8.5432 ; 27.4535] d) [9.7551 ; 30.9649]
b) [16.2435 ; 32.5430] e) None of the above, specify __________
c) [2.4387 ; 7.7412]

HINT:
( − 1) ( − 1) (25 − 1)4 (25 − 1)4
95% = ; = ; ≈ [9.7551 ; 30.9650]
  39.3641 12.4011

Page 1 of 10
Vietnam National University – HCMC Semester 01, Academic year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business
3. Calculate minimum sample require for the amount of money customer spend when coming to
BigC. Known that population standard deviation = 400,000 VND and the amount of
money is spent is within 20,000VND. (assumed population is normal and = 95%)
(3 points)
a) 1536 d) 1536.584
b) 1537 e) None of the above, specify __________
c) 1535.584

HINT:
(400,000)
= / = (1.96) = 1,536.64 ≈ 1,537
(20,000)

4. At Vung Tau beach, the authority claims that the average amount of money a Vietnamese
spend is less than 1,200,000 VND/per. However WLCP (we love cheap price) claims that the
average amount is more than 1,200,000 and decide to make a sample to test. The sample size
is 40 and yield ̅ = 1,300,000 VND/per. Assumed = 50,000 VND and = 5%. Can
the WLCP reject authority claim (3 points)
a) YES REJECT c) Not enough information
b) NO CANNOT d) None of the above, specify __________

HINT: : ≤ 1,200,000; : > 1,200,000

̅− 1,300,000 − 1,200,000
= = ≈ 12.6491
/√ 50,000/√40

Since the test statistic value is relatively large, we can strongly reject the null hypothesis.

5. Vina Banana believe that the amount of banana in a tree is normally distributed with mean =
500 banana/tree and standard deviation = 80. However, due to Global Warming, scientist
says that productivity of banana tree is not and make a sample of 16 trees and have ̅ = 535
for testing their claim. Should scientist reject the claim of Vina Banana? (3 points)
a) Reject with = % d) CANNOT SOLVE
b) Reject with = 5% e) None of the above, specify __________
c) Reject with = 1%

HINT: : = 500; : ≠ 500

̅− 535 − 500 − = 2 ( > 1.75) = 0.08 < 0.1


= = ≈ 1.75
/√ 80/√16

We can reject the null hypothesis at 0.1 level of significance

Page 2 of 10
Vietnam National University – HCMC Semester 01, Academic year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business
6. Toyota engine is assumed to use less than 5 litters for 100km. Toyota users want to test that
claim and make a sample of 25, get the result of 5.35 litters for 100 km with sample standard
deviation s = 0.8 litter for 100 km. Can Toyota user reject the Toyota producer’s claim?
(assumed population is normal) (3 points)
a) Reject with α = 10% d) CANNOT SOLVE
b) Reject with α = 5% & 10% e) None of the above, specify __________
c) Reject with α = 1%, 5% & 10%

HINT: : ≤ 5; : > 5
̅− 5.35 − 5
= = ≈ 2.1875
/√ 0.8/√25

At = 0.01, = , . = 2.492
At = 0.05, = , . = 1.711
At = 0.1, = , . = 1.318

Hence, we can reject the null hypothesis at 0.1 and 0.05 levels of significance.

Use the following data to answer the next 3 questions:


In a study by the United Dairy Industry Association, 42% of 655 persons age 35 – 44 said they
seldom or never ate cottage cheese. For 455 individuals age 45 – 54, corresponding percentage
was 34%. A market research company conduct a hypothesis test to determine if the proportion of
people age 35 – 44, who seldom or never ate cottage cheese is higher than that of people 45 – 54,
Let :
: the proportion of people age 35 – 44, who seldom or never ate cottage cheese
: the proportion of people age 45 – 54, who seldom or never ate cottage cheese

7. The suitable hypothesis for this test would be: (3 points)


a) : − =0 : − ≠0
b) : − ≤ : − >
c) : − ≥0 : − <0
d) : − ≤0 : − >0
e) None of the above, specify __________

8. The test statistic of this test is: (3 points)


a) 2.691 d) 1.645
b) 2.72 e) None of the above, specify __________
c) 2.575

HINT:
+ ̂ + ̂ (655)(0.42) + (455)(0.34)
̂= = = = 0.3872
+ + 655 + 455

Page 3 of 10
Vietnam National University – HCMC Semester 01, Academic year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business
( ̂ − ̂ )−0 (0.42 − 0.34) − 0
= = ≈ 2.691
1 1 1 1
̂ (1 − ̂ ) + 0.3872(1 − 0.3872) 655 + 455

9. Using = 0.01, the conclusion of this test will be: (3 points)


a) > b) < c) =

HINT:
− = ( > 2.7199) ≈ 0.00326. We can reject : − ≤ 0. So >

10. The manufacturer of the X-15 steal-belted radial truck tire claims that the mean mileage the
tire can be driven before the tread wears out is 60,000 miles. The standard deviation of the
mileage is 5,000 miles. The Crosset Tuck Company bought 48 tires and found that the mean
mileage is 59,500 miles. Crosset’s experience is different from that claimed by the
manufacturer at the 0.05 significance level. Is the claim of the X-15 steal-belted radial truck
tire false? (3 points)
a) Yes, it’s false. b) No, it’s true.

HINT: : = 60,000; : ≠ 60,000

̅− 59,500 − 60,000
= = ≈ −0.6928
/√ 5,000/√48

− = 2 ( < −0.6928) ≈ 0.4883 > 0.01

We cannot reject the null hypothesis.

11. In a multiple regression analysis involving 15 independent variables and 200 observations,
SST = 800 and SSE = 240. The coefficient of determination is: (2 points)
a) 0.300 d) 0.700
b) 0.192 e) None of the above, specify __________
c) 0.500

HINT: =1−( / ) = 1 − (240/800) = 0.7

12. A regression model involved 5 independent variables and 136 observations. The critical
value of t for testing the significance of each of the independent variable’s coefficients will
have the degree of freedom of: (2 points)
a) 121 c) 130
b) 135 d) 120

HINT: = − ( + 1) = 136 − (5 + 1) = 130

Page 4 of 10
Vietnam National University – HCMC Semester 01, Academic year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business
13. A multiple regression model has: (2 points)
a) only one independent variable c) more than one independent variable
b) more than one dependent variable d) at least 2 dependent variables

14. In a multiple regression analysis SSR = 1,000 and SSE = 200. The F statistic for this model is
(2 points)
a) 5.0
b) 1,200
c) 800
d) Not enough information is provided to answer this question

15. In a regression model involving 30 observations, the following estimated regression equation
was obtained: = 17 + 4 − 3 + 8 + 8 . For this model SSR = 700 and SSE = 100.
The conclusion is that the: (2 points)
a) model is not significant c) slope of is significant
b) model is significant d) slope of is significant

HINT:

SS df MS F-ratio
Regression 700 4 175 43.75
Error 100 25 4

Since, the test statistic value is too large, we can strongly reject at most of levels of
significance. It means that there is statistical evidence that the model is significant.

Page 5 of 10
Vietnam National University – HCMC Semester 01, Academic Year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business

PART II: (Please answer on ANSWER SHEET)


Question 01: (10 points)
A manufacturer of cereal is considering three alternative box colorsred, yellow, and blue. To
check whether such consideration has any effect on sales, sixteen stores of approximately equal size
are chosen. Red boxes are sent to six of these stores, yellow boxes to five others, and blue boxes to
the remaining five. After a few days, a check is made on the number of sales in each store. The
results (in tens of boxes) shown in the table were obtained.

RED YELLOW BLUE


43 52 61
52 37 29
59 38 38
76 64 53
61 74 79
81

ANOVE TABLE

Source of Variation SS df MS F
Treatment
Error 277.54

Total 3,948.94

Complete the analysis of variance table, and test the null hypothesis that the population mean sales
levels are the same for all three box colors.

Answer to Question 01 (Chapter 09: Analysis of Variance [ANOVA])

Assume normally distributed populations, independent random samples, and equal variances.

: = =
: ( = , , )

Source of Variation SS df MS F
Treatment 340.94 2 170.47 0.6142
Error 3608.00 13 277.54

Total 3948.94 15

= 0.6142 ; = , , . = 3.8056

Since < , we cannot reject the null hypothesis at 0.05 level of significance. It means that
the population mean sales levels are the same for all three box colors at = 0.05

Page 6 of 10
Vietnam National University – HCMC Semester 01, Academic Year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business

Question 02: (10 points)


An insurance company wanted to determine the importance of price as a factor in choosing a
hospital in that region. A random sample of 450 consumers were asked to select “not important,”
“important,” or “very important” as an answer. Respective numbers selecting these answers were
142, 175, and 133. Test the null hypothesis that a randomly chosen consumer is equally likely to
select each of these three answers.

Answer to Question 02 (Chapter 14: Chi – Square Testing for Goodness-of-Fit)

( ) not important, ( ) important, ( ) very important.

: The three answers are equally likely; that is, the probabilities of choosing any of the three
answers are equal: = = = 1/3
: Not all three answers are equally likely; that is, the probabilities of choosing any of the three
answers are not equal.

Expected counts:
1
= = = = 450 ×
= 150
3
( − ) (142 − 150) (175 − 150) (133 − 150)
 = = + + = 6.52
150 150 150

At = 0.05,  = , . = 5.99147

Thus, at 0.05 level of significance, we can reject the null hypothesis. It means there is statistical
evidence that not all three answers are equally likely.

Question 03: (10 points)


Data on the marital status of men and women ages 20 to 29 were obtained as part of a national
survey. The results from a sample of 350 men and 400 men follow. These data are representative of
results published in a daily newspaper.
Marital Status
Gender
Never Married Married Divorced
Men 234 106 10
Women 216 168 16

Using  = 0.01, test for independence between marital status and gender. What is your conclusion?

Answer to Question 03 (Chapter 14: Chi – Square Testing for Independence)

H : The marital status and the gender are independent of each other.
H : The marital status and the gender are not independent of each other.

Page 7 of 10
Vietnam National University – HCMC Semester 01, Academic Year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business

CONTINGENCY TABLE

Marital Status
Gender Total
Never Married Married Divorced
234 106 10 350
Men
(210) (127.87) (12.13)
216 168 16 400
Women
(240) (146.13) (13.87)
Total 450 274 260 750

O −E (234 − 210) (106 − 127.87) (16 − 13.87)


 = = + +⋯+ ≈ 12.86
E 210 127.87 13.87

At = 0.01,  = , .
= 9.21034

Hence, at 0.01 level of significance, we can reject the null hypothesis. It means that there is
statistical evidence for independence between marital status and gender.

Question 04: (10 points)


A student used multiple regression analysis to study how family spending (Y) is influenced by
income (X1), family size (X2), and addition to savings (X3). A random sample of 15 families were
selected. The variables Y, X1, and X3 are measured in thousands of dollars. The following results
were obtained.

ANOVA TABLE
Source SS df MS F
Regression
Error 2.6218
Total

COEFFICIENTS TABLE
Predictors Coefficients Standard Error
Constant 0.0136
X1 0.7992 0.074
X2 0.2280 0.190
X3 -0.5796 0.920

Coefficient of determination = 0.946.

a. Carry out a test of whether Y is significantly related to the independent variables. Use a 0.05
level of significance.
b. What can you say about the strength of this relationship?
c. Carry out the tests of the significance of individual regression parameters.
d. Write out the estimated regression equation for the relationship between the variables.

Page 8 of 10
Vietnam National University – HCMC Semester 01, Academic Year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business

Answer to Question 04 (Chapter 11: Multiple Regression)

a. = /(1 − ) = 2.6218/(1 − 0.946) ≈ 48.5519

ANOVA TABLE
Source SS df MS F
Regression 45.9301 3 15.31
Error 2.6218 11 0.2383 64.2345
Total 48.5519 14

Since the test statistic value is too large, we can strongly reject the null hypothesis at most of
levels of significance. It means that there is statistical evidence for the linear regression
relationship between the dependent variable and at least one of the independent variables. Or, Y
is significantly related to the independent variables.

b. With coefficient of determination = 0.946, we can conclude that 94.6% of the variation in the
dependent variable, family spending, is explained by the combination of the three independent
variables, income, family size, and savings.

c.

COEFFICIENTS TABLE
Predictors Coefficients Standard Error t
Constant 0.0136
X1 0.7992 0.074 10.80
X2 0.2280 0.190 1.20
X3 -0.5796 0.920 - 0.63

H : β = 0 H : β = 0 H : β = 0
H : β ≠ 0 H : β ≠ 0 H : β ≠ 0

At = 0.05, the critical value: ± =± , / =± ( ), / =± , . = ±2.201

With tests of the significance of individual regression parameters, only the independent variable,
X1, is significant and has explanatory power with respect to the dependent variable Y. And X2
and X3 are non-significant since we cannot reject the null hypothesis that their slope parameters
are zero and have no explanatory power. Hence, they should be removed from the regression
equation. And we should run the regression equation again.

d. The estimated regression equation for the relationship between the variables:

= 0.0136 + 0.7992 + 0.2280 − 0.5796

Question 05: (10 points)


A study is undertaken to determine how consumers react to energy conservation efforts. A random
group of 60 families is chosen. Their consumption of electricity is monitored in a period before and
a period after the families are offered certain discounts to reduce their energy consumption. Both
periods are the same length. The difference in electric consumption between the period before and
the period after the offer is recorded for each family. Then the average difference in consumption
and the standard deviation of the difference are computed. The results are = 0.2 kilowatt and

Page 9 of 10
Vietnam National University – HCMC Semester 01, Academic Year 2013 - 2014
INTERNATIONAL UNIVERSITY Course: Statistics for Business
= 0.1 kilowatt. At = 0.01, is there evidence to conclude that conservation efforts reduce
consumption?

Answer to Question 05 (Chapter 08: The Comparison of Two Populations)

Assume normal distribution. : ≤ 0; : >0

− 0.2 − 0
= = ≈ 15.4919
/√ 0.1/√60

= 2.33

At 0.01 level of significance, we can reject the null hypothesis. It means there is statistical
evidence to conclude that conservation efforts reduce consumption.

Question 06: (10 points)


The national safety Council reported that 52 percent of American turnpike drivers are men. A
sample of 300 cars travelling eastbound on the Ohio Turnpike yesterday revealed that 170 were
driven by mean. At the 0.01 significance level, can we conclude that a larger proportion of mean
were driving on the Ohio Turnpike than the national statistics indicate?

Answer to Question 05 (Chapter 07: Hypothesis Testing)

Assume large population. : ≤ 0.52; : > 0.52

̂− (170/300) − 0.52
= = ≈ 1.6179
(1 − ) 0.52(1 − 0.52)
300

= 2.33

At 0.01 level of significance, we cannot reject the null hypothesis. Hence, there is no evidence
to conclude that a larger proportion of mean were driving on the Ohio Turnpike than the
national statistics indicate.

------ THE END ------

Page 10 of 10

You might also like