1

SBIAHU AMS 315, Fall 2022 Hongshik Ahn

FINAL EXAM November 29, 2022

NAME
STUDENT ID #

Instructions:

• Distribution tables are given on the last 4 pages.

• Answers should be given clearly.

• Justify your answer to obtain full credit.

• A hand calculator is allowed.

• Your cell phone cannot be used as a calculator.

• The maximum possible score is 140.

• Partial credit policy: Questions left blank, or with work that clearly does not lead to the
solution, will receive 0 points. Partially correct work (or work clearly in the right direction)
will receive partial credit.

Academic Integrity Statement:

Academic integrity is expected of all students at all times, whether in the presence or absence
of members of the faculty. Understanding this, I declare that I shall not give, use, or receive
unauthorized aid in this examination. I further understand that the consequences of academic
dishonesty related to an examination are an automatic grade of 0 for the course.

SIGNATURE
 2

1. (13 points: 10, 3) While estimating a population proportion using a large sample, one reports
that the estimate of p is p̂ = .32 and its 90% error margin is .08.

(a) Find a 95% confidence interval for the population proportion.

(b) Find the sample size to satisfy the error margin.

2. (14 points: 9, 3, 2) An educator wishes to test if more than 30% of the college football players
graduate college in four years.

(a) If 19 out of a random sample of 48 players graduated in four years, what does the test
conclude? Test using α = .05.
(b) Find the p-value of the test.
(c) Suppose a further study shows that, in fact, 35% of the college football players graduate
in four years. Refer back to the decision in part (a), what did your analysis lead to?
(i) Type I error (ii) Type II error (iii) A correct decision

3. (13 points: 10, 3) An independent bank, concerned about its customer base, decided to conduct
a survey of bank customers. Out of 505 customers who returned the survey form, 258 rated
the overall bank services as excellent.

(a) Test, at level α = .10, the null hypothesis that the proportion of customers who would
rate the overall bank services as excellent is .46 versus a two-sided alternative.
(b) Calculate the p-value and comment on the strength of evidence.

4. (15 points) A shipment of assorted nuts is labeled as having 45% walnuts, 20% hazelnuts, 20%
almonds, and 15% pistachios. By randomly picking several scoops of nuts from this shipment,
an inspector finds the following counts. Could these findings be a strong basis for an accusation
of mislabeling? Test at α = 0.01.

Walnuts Hazelnuts Almonds Pistachios Total

Counts 95 70 33 42 240

5. (15 points) In a genetic study of chromosome structures, 132 individuals are classified according
to the type of structural chromosome aberration and carriers in their parents. The following
counts are obtained.

Carrier
Type of aberration One parent Neither parent Total
Presumably innocuous 27 20 47
Substantially unbalanced 36 49 85
Total 63 69 132

Test the null hypothesis that type of aberration is independent of parental carrier. Use α = .05.
 3

6. (35 points: 8, 3, 3, 4, 5, 5, 7) A study was conducted with sampling units equal to individual
Douglas-fir trees in the Western Cascades, Oregon. For each tree in the sample, two variables
were measured: X = tree age in years, and Y = 5-year height growth (cm). Use the computer
output given below and the following numerical summaries to answer the questions.
n = 16, x = 12.31, y = 266.875, sxx = 175.44
The regression equation is: Y = 4.4 + 21.3X.

Predictor Coef Stdev t-ratio

Constant 4.35 49.51 0.09
Tree age in years 21.322 3.883 5.49

s = 51.43 R2 = 68.3%

(a) Determine the following values from the computer output.

The estimated slope of the regression line:
The estimated intercept of the regression line:
The estimated standard deviation of the tree age in years:
The estimate of the standard deviation around the true regression line:
(b) Which of the following is the best interpretation of the value of R2 ?
i. 68.3% of the observed variation in 5-year height growth can be explained by a linear
relationship with tree age.
ii. Adjusting for tree age results in a 68.3% decrease in the standard deviation of 5-year
height growth.
iii. For approximately 68.3% of the tree in the population, ŷ is a better predictor of y
than y.
iv. Tree age can be used to predict 5-year height growth for 68.3% of the trees.
(c) What is the prediction for the amount of 5-year height growth for a 10-year-old tree?
(d) Find a 95% confidence interval for the slope of the population line.
(e) Find a 90% confidence interval for the mean 5-year height growth of all 10 year-old trees.
(f) Find a 90% prediction interval for the actual 5-year height growth of a particular 10
year-old tree.
(g) Conduct a hypothesis test if the goal is to show that there is a relationship between X
and Y . Use α = .05.
 4

7. (35 points: 6, 4, 5, 3, 3, 4, 4, 3, 3) We are interested in studying on systolic blood pressure

y in relation to weight x1 and age x2 in a class of males of approximately the same height.
From 14 subjects preselected according to weight and age, data are obtained. The regression
results are summarized in the computer output given below.

The regression equation is ŷ = β̂0 + β̂1 x1 + β̂2 x2 = −71.23 + 1.097x1 + 0.479x2 .

Predictor Coef Stdev t-value p-value
Constant -71.2278 24.4734 (a) 0.014
x1 1.0966 0.1265 (a)
x2 0.4789 0.1187 (a)

The ANOVA table is given below.

df SS MS F value p-value
x1 1 1141.10 (d) (e)
x2 1 277.16 (d) (e)
Residuals 11 187.24 (d)

(a) Find the t-values of the coefficient estimates of the constant, x1 , and x2 .
(b) Find a 95% confidence interval for a coefficient β1 .
(c) Do these data strongly indicate that the blood pressure increases as a person gets older?
Test at α = 0.01.
(d) Find the three MS values in the above ANOVA table.
(e) Find the two F values in the above ANOVA table.
(f) Fill out the following ANOVA table.
df SS MS F value
Regression ( ) ( ) ( ) ( )
Error ( ) ( ) ( )
Total ( ) ( )
(g) Find the R2 .
(h) Estimate the residual standard error.
(i) Is this regression model significant at level α = 0.05? Answer the question by testing
H0 : β1 = β2 = 0 vs. H1 : either β1 = 0 or β2 = 0.
5
6
 7

Table A.5 Chi-square Distribution Table (1)

8