Homework 14

Due Wednesday, December 11, 11:59pm

STAT 400, Fall 2024, D. Unger

You may use R code and output as an option for showing work. It can be copied and pasted
to your electronic file (LaTeX, notewriter, etc.) or copied by handwriting to a pen and paper
submission that you later scan.

Exercise 1
We wish to create a two-sided test at the α = 0.05 level of significance where H0: µ = 75 and
H1: µ ≠ 75. Suppose we are given σ = 5 and n = 25.
As seen in lecture, the first two steps for finding the power of a test are the same as
conducting the test itself.
1. State the parameter of interest as well as the two competing hypotheses.
2. State the point estimator for the parameter and its sampling distribution.
The third step for the actual calculation of Power is broken up into the following sub-steps
(a) through (c).
(a) Find the critical value for the test in terms of z.
(b) Find the critical value for the test in terms of 𝑥̅ .
(c) Find the power of this test when µ = 72.
(d) Find the power of this test when µ = 80.

Exercise 2
This revisits an exercise from Homework 12.
A trucking firm knows that its mean weekly loss due to damaged shipments is historically
at most $1800. However, there are concerns about increased losses due to damages. Half a
year (26 weeks) of operation shows a sample mean weekly loss of $1921.54 with a sample
standard deviation of $249.39.
The trucking firm chooses to use an α = 0.10 level of significance.
(a) Find the critical value for the test in terms of z.
(b) Find the critical value for the test in terms of 𝑥̅ .
(c) Find the power of this test when µ = $1921.54 (the observed sample mean).
(d) What alternative value of µ would yield a power of 0.90?

Exercise 3
A home builder is testing concrete compressive strength in advance of an upcoming
housing development. Under the null hypothesis, the mean strength is at least 3000psi as
advertised by the concrete manufacturer. Of course, the builder will have a problem on
their hands if the concrete tests at significantly less than that. Assume that the strengths
are normally distributed with population standard deviation of 500 psi.
(a) State the null and alternative hypotheses of the builder’s test. Given a sample of size 10,
define a rejection region in terms of mean strength, 𝑋, under the one-sided alternative
hypothesis at significance level α = 0.05.
(b) Based on the result from part a, write a power function κ(μ) for this test of the concrete.
That is, write a function such that when you input a value of μ, the output is the power at
that value of μ. Also, what is the domain of that function, also known as the parameter
space of μ? And separate from that, for what values of μ, does it actually make practical
sense to calculate power?
(c) Using your new power function κ(μ), calculate the power of the test at the alternative
true mean strength of 2600 psi. Explain the calculated value of power in the context of the
problem.
(d) Based on the result from part (c), what is the probability of a Type II error when the
true mean compressive strength is 2600psi? Explain this probability value in the context of
the problem.

Exercise 4
Over a long period of time at a suburban restaurant, in addition to the glass of water that all
customers get, 60% of the customers ordered coffee, 20% ordered tea, 15% ordered soda,
and 5% ordered milk. The owners opened a new restaurant in the business section of town,
where they think that the drinking preferences of the customers may be different. In a
random sample of 300 customers, 150 ordered coffee, 78 ordered tea, 51 ordered soda, and
21 ordered milk.
Determine at a 1% level of significance whether drinking patterns are different in town
than at the suburban restaurant.
(a) Define the parameters of interest as well as the null and alternative hypotheses.
(b) Find the value of the test statistic for this test.
(c) Define the critical region (i.e., the rejection region) for this test, and make a decision.
Also include your conclusion in the context of the scenario.
(d) In the post-hoc analysis of a goodness-of-fit test, we examine the individual chi-square
terms that sum together to formulate the chi-square test statistic. The largest of those
terms reveal where the significant differences occur leading to the rejection of the null
hypothesis. Terms near the value of the degrees of freedom are suspect, and those above it
are truly noteworthy.
What is the value for degrees of freedom in this test, and which groups have chi-square
terms larger than that value?

Exercise 5
A biostatistician randomly samples 400 nucleotides from the human genome. (Consider
only a single strand.) Does the occurrence of adenine, thiamine, cytosine, and guanine (also
known as A, T, C, and G) in this sample suggest that nucleotides in the human genome
follow a uniform distribution?
Use a χ2 test with a significance level of α = 0.01 to test the following hypotheses:
H0: pA = pT = pC = pG = 0.25 vs.
H1: At least one of pA, pT, pC, or pG is not equal to 0.25
The data collected is as follows.

A T C G

Frequency 90 85 110 115

(a) Calculate the value of the appropriate test statistic.

(b) State the critical value for this test, your statistical decision, and a concluding statement
in the context of the problem.
Scientific Note: Even if A, T, C, and G follow a uniform distribution in the human genome,
that does not mean that nucleotides occur randomly! Even with a uniform distribution,
certain patterns can and do occur.
Exercise 6
In a random sample of 1,000 voters in Champaign-Urbana, each individual was asked to
name the issue that is most important to them in the upcoming 2024 presidential election.
The individuals were also classified by party affiliation. The results were as follows.

Most Important Issue

Party Affiliation Economy Terrorism Social Issues Climate Change Total

Democrat 193 135 112 60 500

Republican 176 129 64 31 400

Independent 31 36 24 9 100

Total 400 300 200 100 1000

We wish to test whether political party affiliation and the most important issue are
independent of one another among Champaign-Urbana voters.
(a) Provide the null and alternative hypotheses for this test. Calculate the value of the
appropriate test statistic.
(b) State the critical value for this test, and your statistical decision using a 5% significance
level. Provide a concluding statement in the context of this situation.