STAT2012 Statistical Tests

General Information

This page contains information on the Intermediate Unit of Study STAT2012 Statistical Tests.

This unit is offered in Semester 2.

For further information on Intermediate Mathematics and Statistics, refer to the Intermediate Handbook. In particular, see the STAT2012 handbook entry for further information relating to STAT2012.

You may also view the description of STAT2012 in the central units of study database.

Credit point value: 6CP.
Classes per week: Three lectures, one tutorial and one computer laboratory session.

Email enquiries about STAT2012 may be sent to STAT2012@maths.usyd.edu.au.

Students: Please give your name and SID when emailing us. Anonymous emails will not be replied to.

Links to online resources for this unit of study will be added here as the material becomes available.

Course information

New announcements

My consultation time for exam. is Nov 13 (Fri) 3-5pm, Nov 16 (Mon) 3-5pm, Nov 18 (Wed) 3-5pm and Nov 19 (Thur) 3-5pm.
Please complete the unit of study survey online.
You should read all informations on this website regularly. In particular, you should read Feedback and right concepts under computer practical to help you improve your computer practical performance. Seek help from your friendly tutors if necessary.
Lecture recordings are available from Blackboard.
Course discussion

Lecture notes

Tutorials

Use of R and RStudio

Introduction to R - 1
Introduction to R - 2
Download R software from internet
click: Windows, base, R-2.9.1-win32.exe, Run, Run, OK (English), Next>, Next>, Next>(unless you change the directory), ...
General Latex commands: Maths mode, Maths symbol, Superscripts & subscripts, Greek letters, Fraction & Roots

Computer practicals

You should attend the computer class and work out the report during the class, NOT earlier or later. You can't simply submit your report at the start of class. Also, your report will NOT be marked if you skip class without approval. If you need extra time after the end of your lab class, you may stay behind or move to another lab to complete your work.
Remember to write down your name, SID and the session, say Monday, in the report. Otherwise, your marks may not be recorded.
Some of you find great difficulty in typing the H0, H1 and other necessary mathematical symbols in your computer report. Please note that the file Latex commands for writing comments under Use of R and process provides you some hints.
If you get error message for your plots in which you think your commands are right, it might be due to the system being sweave instead of knitr.
To change back to knitr, click Tools, Global Options, Sweave (the 6th option on the left). Ensure that the first box says using: knitr. Then click Ok to save. At the top of your file after \begin{document}, remove \SweaveOpts{concordance=TRUE}
To run in sweave, remove < < include=FALSE > = opts_knit$set(concordance=TRUE) @ and set < < fig=TRUE > > =
Note: When p-value>0.05, you should say "insufficient evidence against H0" or "the data are consistent with H0". Don't say "the data are against H1", "sufficient evidence supporting H0" nor "H0 is true". We can't find any evidence favoring or supporting H0 as we assume H0 is true in calculating the p-value. Also you never know whether H0 or H1 is true just based on a sample. We aim to find evidence against H0. So when p-value<0.05, you should say "sufficient (or strong evidence) against H0". Don't say "insufficient evidence supporting H0".
From week 9, no template is available. You need to create your own template, as in the computer exam. on week 13.
Creating reports using Rstudio
Latex commands for writing comments
Feedback and right concepts
Week 1: Practical 1 Template Solution
Week 2: Practical 2 Template Solution
Week 3: Practical 3 Template Solution Survey data
Week 4: Practical 4 Template Solution
Week 5: Practical 5 Template Solution
Week 6: Practical 6 Template Solution
Week 7: Practical 7 Simplified template Solution itch data
Week 8: Practical 8 Simplified template Solution range data auto data
Week 9: Practical 9 Solution price data
Week 10: Practical 10 Solution fuel data
Week 11: Practical 11 Solution
Week 12: Practical 12 Solution
Week 13: Computer exam.

Assignments

You may write or type your assignment. You need to show all steps and you are encouraged to include R outputs to check your answers. You may just staple the R output together with your hand written or typed assignment. However just R output even showing all steps is not acceptable cause I want you to work out the answer by youself as if in the quiz or exam. Include a Cover sheet and put your assignment into the assignment box for STAT2012 just outside my office in Carslaw Rm 817 on or before 5:00 pm of the due date. You may also hand it to me during the lectures or slide under my door.
For assignment 2 Q2 on Bonferroni test, read P.157 of the lecture note. It calculates the test statistic for 2-sample t-test for one pair of treatments. Bonferroni test refers to simultaneously testing for all pairs of treatments. Hence you simply repeat the calculation for all 4 choose 2=6 pairs of treatments.
Assignment 1: Due Aug 26 (5%) Solution
Assignment 2: Due Oct 14 (5%) Solution

Quiz

The quiz will be held on Sept 16, 9:00am lecture in Carslaw 273 and 173 and it accounts for 5%. There will be 10 MC questions on materials from Week 1 to 7. Since problems on 2-way ANOVA and Friedman test only appear in Wk 8 tutorial, they are excluded. Statistical table and formulae table are provided. You don't need to bring your own formulae sheets and statistical tables but remember to bring your own calculator and student ID for checking.
Practice quiz
Notes for quiz
Mark distribution

Computer practical examination

The template for computer exam. will NOT be provided. However previous simplified templates for week 7 & 8 can be adopted with modification.
Some of you find great difficulty in typing the H0, H1 and other necessary mathematical symbols in your computer report. Please note that the file Latex commands for writing comments under Use of R and process provides you some hints.
Note for practical and exam.: When p-value>0.05, you should say "insufficient evidence against H0" or "the data are consistent with H0". Don't say "the data are against H1", "sufficient evidence supporting H0" nor "H0 is true". We can't find any evidence favoring or supporting H0 as we assume H0 is true in calculating the p-value. Also you never know whether H0 or H1 is true just based on a sample. We aim to find evidence against H0. So when p-value<0.05, you should say "sufficient (or strong evidence) against H0". Don't say "insufficient evidence supporting H0".
Note for computer exam.

About the School

Research

For Current Students

For Future Students

Internal Pages