Option 5 2018 Final Exam

Question 1
a) If you made 30 estimates of population means using a 90% confidence interval, how many of
those intervals would you expect to contain the true mean?
b) You want to be 95% sure that at least 4 of you next 5 intervals will contain the true mean.
How wide should you make your confidence interval?

Question 2
A large population, X, has proportion p=0.53, and 𝑋 is the proportion of a random sample of size
n drawn from the population of X. Find the least value of n such that P(𝑋>0.5)>0.95

Question 3
A farmer sells bags of potatoes which he states have a mean weight of 7kg. An inspector, however,
claims that the mean weight is less than 7kg. In order to test this claim, the inspector takes a
random samples of 12 of these bags and determines the weight, x kg, of each bag. He finds that
∑ 𝑥 = 83.64 ∑ 𝑥 2 = 583.05
a) State an appropriate assumption that allows you to test the inspector’s claim
b) Find unbiased estimates of 𝜇 and 𝜎 2
c) Text, at the 10% significance level, the claim that the mean weight is less than 7kg.

Question 4
Unbiased estimates of the mean and variance of a population, based on a random sample of 19
observation are 4.7 and 1.64 respectively. Another random observation of 2.0 is obtained. Find
new unbiased estimates of the mean and the variance with this extra observation taken into
account.

Question 5
Given random variables X and Y, where 𝜇𝑥 = 4 and 𝜎 2 𝑥 = 2.25 and Y=2X+1, find Cov(X,Y)

Question 6
12 high school students are measured, and their heights and ages recorded below:
Age, x(in years) 12 12 12 13 13 13 14 14 14 15 15 15
Height, y(in cm) 143 128 144 157 151 163 164 161 168 174 171 181
Answer the following questions using your graphics calculator - no working is required.
a) Find the value of the product moment correlation coefficient for this data and state the nature
of any correlation.
b) Find the equation of the regression line for:
i) Y on X
ii) X on Y
c) Predict the age (to the nearest month) of a student who is 154cm tall.
d) Do you think the linear model will continue to be appropriate regardless of how large or small x
becomes? Justify your answer.
Question 7
A statistician chooses 27 randomly selected dates, and when examining the occupancy records of
a particular motel for those dates, finds a standard deviation of 5.86 rooms rented. If the number
of rooms rented is normally distributed, find the 95% confidence interval for the population
standard deviation of the number of rooms rented.
(𝑛−1)𝑠2
[Use the fact that 𝜎2
~χ2 (𝑛 − 1) to construct this confidence interval by hand.]

Question 8
Glados boasts that she can jump at least 40cm longer than Chell can. Chell disputes this-and so
they each make several jumps to check how far apart their averages are.
Chell’s results are 𝑛𝑥 = 22, ∑ 𝑥 = 6182 , ∑(𝑥 − 𝑥)2 = 4980
Glados’ results are 𝑛𝑦 = 28, ∑ 𝑦 = 9156 , ∑(𝑦 − 𝑦)2 = 9442 (in cm)
Test whether there is evidence at the 5% level support Glados’ claim. State any assumptions you
make.

Question 9
An arcade owner is asked how often he expects the Ticket Jackpot machines to pay out. He states
confidently that they are programmed to go off an average of 3 times a week.
Data from a total of 64 machine-weeks (all machines assumed to have the same distribution) are
collected, and the following results produced:
Number of Jackpots 0 1 2 3 4 5

Number of Weeks 9 15 18 13 7 2
No weeks were recorded with a machine jackpotting more than 5 times.
Create an Observed/Expected frequency table and carry out a χ2 goodness-of-fit test of the
model 𝑃𝑜(3) at the 1% level. Calculate a value of 𝜆 that would be more likely to pass this test.

Question 10
It’s believed that 30% of those born in Australia who are over 65 have blue eyes. To test this
hypothesis, 68 senior citizens are approached and their eye colour recorded.
a) When performing this hypothesis test at the 10% level, how many blue-eyed people would need
to accept/reject H𝑜?[Be careful about＞vs≥and/or＜vs≤signs]
b) It turns out that the true proportion in this population was only 18%. Calculate the probability
of making a type II error.
Question 11
a) Fill out the following Cumulative Probability Table for a binomial, X, with n=10 and p=0.5
X 0 1 2 3 4 5 6 7 8 9 10
P(X≤x)
b) Let Y be a Normal approximation X. State the distribution of Y below and write down P(Y≤x) in
the space underneath the table above.
c) What is the largest absolute difference between X and Y in cumulative probabilities？
d) The maximum error for any given value of p is known to be proportional to the reciprocal square
𝑎
root of n-that is to say, Errormax= 𝑛, where a is a constant. Using this, find the lowest value of n for
√

which Errormax<0.005
e) Rather than increasing n, we can increase the accuracy of our approximation by applying a
continuity correction. What is the largest error in our original distribution if we apply this correction?

Question 12
Let 𝑋1 , 𝑋2, … , 𝑋81 be independent identical distributions, each with expected value
𝜇 = 𝐸(𝑋𝑖 ) = 5 and variance 𝜎 2 = 𝑉ar(𝑋𝑖 ) = 4. Evaluate P(𝑋1 + 𝑋2 + ⋯ + 𝑋81 > 369)

Question 13
You want to estimate the size of a class where you cannot the students directly. You know the
students are numbered from 1 to n, where n is the number of students in the class. You call out
three students at random and ask for their numbers, which turns out to be 1,3,7. Find the
maximum likelihood estimate for n, justifying your answer appropriately.

Question 14
Consider the sequence of random variables 𝑋1 , 𝑋2 , ⋯ where the pdf of 𝑋𝑛 is equal to
𝑛−1 1 1
, − <𝑥<
2 𝑛 𝑛
𝑓𝑛 (𝑥) = 1
, 𝑛 <𝑥 <𝑛+1
𝑛
{ 0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

A) What is 𝐸(𝑋𝑛 )?
B) What is the variance of 𝑋𝑛 ?
C) What is the variance of 𝑋𝑛 as 𝑛 → ∞?