Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
MAST30025 Linear Statistical Models Assignment 3
Submission deadline: Friday May 27, 5pm
This assignment consists of 4 pages (including this page) with 5 questions and 47 total marks
Instructions to Students
Writing
This assignment is worth 7% of your total mark.
You may choose to either typeset your assignment in LATEX, or handwrite and scan it to
produce an electronic version.
You may use R for this assignment, including the lm function unless otherwise specified.
If you do, include your R commands and output.
Write your answers on A4 paper. Page 1 should only have your student number, the
subject code and the subject name. Write on one side of each sheet only. Each question
should be on a new page. The question number must be written at the top of each page.
Scanning and Submitting
Put the pages in question order and all the same way up. Use a scanning app to scan all
pages to PDF. Scan directly from above. Crop pages to A4.
Submit your scanned assignment as a single PDF file and carefully review the submission
in Gradescope. Scan again and resubmit if necessary.
MAST30025 Linear Statistical Models Assignment 3 Semester 1, 2022
Question 1 (7 marks)
Let X =
[
1 2 1
2 1 1
]
and let A = XTX.
(a) Calculate A.
(b) Find a conditional inverse Ac such that r(Ac) = 1, or show that no such conditional inverse
exists.
(c) Find a conditional inverse Ac such that r(Ac) = 2, or show that no such conditional inverse
exists.
(d) Find a conditional inverse Ac such that r(Ac) = 3, or show that no such conditional inverse
exists.
Question 2 (11 marks)
Consider a one-way classification model
yij = µ+ τi + εij
for i = 1, 2, 3 and j = 1, 2, . . . , ni. The following data is collected:
Factor level: A B C
ni 12 8 16
Mean response: 11.3 8.4 10.2
We are also given s2 = 4.9.
For this question, you may not use the lm function in R.
(a) Calculate a 95% confidence interval for τA − τB.
(b) Calculate the F -test statistic for the hypothesis τA = τB = τC , and state the degrees of
freedom for the test.
(c) Test the hypothesis H0 : τC − τB ≥ 2 against H1 : τC − τB < 2 at the 5% significance level.
(d) Suppose the above data is collected through a completely randomised design with total
sample size n = 36. Does this design minimise 2var (τˆA − τˆC) + var (τˆB − τˆC)? If not,
what is the optimal allocation for nA, nB, and nC?
Page 2 of 4 pages
MAST30025 Linear Statistical Models Assignment 3 Semester 1, 2022
Question 3 (9 marks)
Consider the two-factor model with interaction
yij = µ+ τi + βj + ξij .
Suppose that there are a and b levels of the factors respectively. Now consider the set of
equations
ξij − ξ1j − ξi1 + ξ11 = 0, i = 2, . . . , a, j = 2, . . . , b.
(a) Show that the equations are not redundant.
(b) Show that these equations are equivalent to the hypothesis of no interaction.
(c) Thereby calculate the rank of the hypothesis of no interaction.
(d) Show that the hypothesis is testable, provided there exists at least one sample from each
combination of factor levels.