Statistics 1 (MATH11400) unit website Statistics 1 - (MATH11400) - 2020

Overview

Lectures are taking place on Tuesdays 15-16 (PHYS BLDG 1.11 TYNDALL) and Thursdays 14-15 (PHYS BLDG 1.11 TYNDALL).

There are bi-weekly one hour tutorials, as in teaching block 1, starting in week 14--check your calendar.
R plays an important role in the course and being able to program is definitely a plus on the job market.

Fortnightly homework involves coding using the statistical software package R (embedded in R studio).

There will be a computing assignment to complete during ~~Weeks 19 and 20~~ with a new deadline of 24th April, noon . You should submit your assessment electronically on Blackboard, using the same procedure you have used to submit your R homework so far -- see this document.

R support will be provided fortnightly Statistics 1 – R studio support by Skevi Michael and Tjun Yee Hoh in weeks 13-15-17-19, in FRY BLDG LG.21 PC Statistics 1 – R studio support . Your allocated time slot should appear in your electronic calendar. These sessions will start in week 13 and you are encouraged to attend from the beginning to familiarize yourself with the R studio interface--do not leave this till the last minute.

Additional R support will be provided in weeks 14 and 16 in FRY BLDG LG.21 PC. These are walk-in sessions and you can choose session(s) from the following: Monday 16-17, Tuesday 13-14, Wednesday 12-13and Friday 15-18.

Additional R information can be found below.

Formal details of this course are available on the unit description page.

Drop-in sessions: Tuesdays, right after the lecture. I will be in my office (GA.11) if you would like to ask about the course.

Lecture notes and homework

Paper copies of the lecture notes are distributed at the beginning of the course only, together with the problems sheets .

With n=14,16,18,20 and 22, homework n is to be handed in to your tutor in week n and covers the material of weeks n-1 and n-2 (except for week 14!).

Solutions to the problems sheets will be made available in week n+1.

Lecture notes (with gaps filled in the lectures by me) will be made available at the end of each week.

This should not give you a false sense of security and encourage you to miss lectures. Experience shows that attending lectures is the best way to remain engaged with the material covered in this course.

There are 10 weeks of lectures, followed by revision sessions.

Week 13 (starting 27/1)

Lecture notes: chapter 1 (see below).
Computing drop in session: simple exercises (once you have gone through this document ).
The `Self learn' document is a good way to start to learn R.
Even more R from Tjun Yee: pdf file and Markdown file.

Week 14 (starting 3/2)

Lecture notes: chapter 1 and 2. There is a typo in Example B.5: the probability of the empty set is 0 and that of Omega 1, naturally...
Week 14 problems sheet solutions. Corresponding Markdown file and pdf file.

Week 15 (starting 10/2)

Lecture notes: chapter 3a and 3b.
Computing drop in session.
More R exercises from Skevi -- here is the Markdown file .

Week 16 (starting 17/2)

Lecture notes: chapter 4a and 4b.
Some R code to help understand the notion of sampling distribution of an estimator (right-click to save).
There have been questions about how R will be assessed in the exam. Here are the last two years exams, 2018 and 2019, to give you an idea .
Week 16 problems sheet solutions (with Part B now included). Markdown file and pdf file.

Week 17 (starting 24/2)

Lecture notes: chapter 5a and chapter 5b.
Computing drop in session. R exercises from Skevi -- Markdown file here.

Week 18 (starting 2/3)

Lecture notes: chapter 6a and chapter 6b -- I have re-recorded the latter on Re/Play to practice.
This week the office hour will be on Thursday, after the lecture.
Some R code to help you understand what a confidence interval is, via simulation.
Week 18 problems sheet solutions. R: Markdown file and pdf file.

Week 19 (starting 9/3)

Lecture notes: chapter 7.
Computing drop in session and computing assignment released. The statistical part is being covered this week in the lectures, but you can already start programming the function in Question 4 during this week's computer lab.
Wondering:

why you are being asked to use computers on this statistics course?
where you are on your journey to becoming a data scientist?

Then have a look at this book. Note that while the book will not necessarily help with the exam it can help you develop an understanding of the data science landscape, if that's for you.

Interesting reading on CIs and testing, pointed out to me by Anthony Lee: Hoekstra, R., Morey, R. D., Rouder, J. N., & Wagenmakers, E. J. (2014). Robust misinterpretation of confidence intervals. Psychonomic bulletin & review, 21(5), 1157-1164. This points to the fact that these ideas are far from easy, although the mathematics looks fairly simple. Some time is needed to really understand these concepts.

Week 20 (starting 16/3)

Lecture notes: chapter 8.
Week 20 problems sheet solutions.

Week 21 ~~(starting 23/3)~~ (starting 20/4)

Lecture notes: chapter 9. Recorded lecture available on mediasite.

Week 22 ~~(starting 20/4)~~ (starting 27/4)

Lecture notes: chapter 10 -- 10.5 and 10.6 are not examinable this year, although it is straightforward material if you have understood earlier chapters.
Week 22 problems sheet solutions (together with week 24 for completeness only).
Computing assignment solutions / markdown file .
2019 exam with solutions (note that the marking details were, as always, indicative only).

If you misplace your lecture notes here is a pdf file of the lecture notes, with gaps. I will not provide you with a second set of printed lecture notes.

R support

To get started you should read the following document.

R demos

You can find a set of R demos which illustrate some of the concepts covered in the lecture notes here.

Textbooks

John A. Rice, 1995, Mathematical Statistics and Data Analysis, Duxbury Press, 2nd ed. This is the international student edition. There is also a 3rd edition, but this seems to be very expensive.
A First Course in Probability by S. Ross.
Introductory Statistics with R by Peter Dalgaard.

Statistics 1 - (MATH11400)