MSG830 V25 Statistisk Analys och Experimentplanering
General course information
On this page you can find the preliminary program for the course: lectures, exercise sessions and computer labs. Below the program, you find a description of compulsory assignments, the exam and the project below the program.
Course is given in English, but course literature is in Swedish as a Kompendium. Swedish slides from a previous iteration are also shared here, and you can find a dictionary of technical terms here: Ordbok. Answers to Kompendium exercises are here: Svarar.
For the labs and the project, the statistical software RStudio is needed, which can be downloaded in two steps
Step 1: install R from http://ftp.acc.umu.se/mirror/CRAN/ (Links to an external site.)
Step 2: install RStudio from https://rstudio.com/products/rstudio/download/ (Links to an external site.)
No prior knowledge of programming is required.
For those who do not have their own computer, there is the possibility to use the computers in room MVF22. You must then ensure that you have an account at Chalmers, which you obtain via the Chalmers student portal.
Extra support: the Maths department offers peer tutoring to all students, you can find out more about it here: https://www.lib.chalmers.se/studieresurser/mattesupport-och-laexhjaelp/
Examiner and course leader: Eszter Lakatos (eszter.lakatos@chalmers.se)
Schedule: TimeEdit
Project: an example minimal project report is available here.
! Update !: June re-exam questions with solution are here: MSG830_250610_withsolutions.pdf
Program
| Date | Session | Topic | Material |
| Onsdag 22/1 |
Föreläsning (slides) |
Datatyper, Deskriptiv statistik | Kapitel 1 |
| Torsdag 23/1 |
Föreläsning (slides) |
Sannolikhetsteorins grunder, Kombinatorik, Betingad sannolikhet, Oberoende händelser, Bayes formel | Kapitel 2 |
| Tisdag 28/1 | Övning |
Övningar i Kapitel 1 och 2: |
1.8.1, 2.6.1, 2.6.2, 2.6.3, 2.6.4 (lösningar till kap. 1-2) |
| Onsdag 29/1 | Datorlab | Beskrivande statistik | Lab 1 (Rmd) (html) |
| Onsdag 29/1 |
Föreläsning (slides) |
Diskret slumpvariabel, Fördelning, Väntevärde, Varians, Binomialfördelning | Kapitel 3 |
| Torsdag 30/1 |
Föreläsning (slides) |
Kontinuerlig slumpvariabel, Täthet, Likformig, Normal | Kapitel 4 |
|
Tisdag 4/2 |
Övning |
Övningar i Kapitel 3 och 4: |
3.8.4, 3.8.5, 4.4.2, 4.4.4, 3.8.1, 4.4.6 (lösningar till kap. 3-4) |
| Onsdag 5/2 | Datorlab | Binomial och normalfördelning | Lab 2 Rmd |
| Onsdag 5/2 |
Föreläsning (slides) |
Centrala gränsvärdessatsen, Population och stickprov, Medelfel, Konfidensintervall, Hypotesprövning | Kapitel 5 |
| Torsdag 6/2 |
Föreläsning (slides) |
Hypotesprövning, t-fördelning, t-test | Kapitel 5 & 6 |
| Tisdag 11/2 | Övning | Övningar i Kapitel 5: |
5.6.2, 5.6.3, 5.6.5, 5.6.7, 5.6.4 (lösningar till kap. 5) |
| Onsdag 12/2 | Datorlab | Filinläsning, Paket & CLT | |
| Onsdag 12/2 |
Föreläsning (slides) |
t-test, parat t-test, ickeparametriska test | Kapitel 6 |
| Torsdag 13/2 |
Föreläsning |
ANOVA, Multipla jämförelser, Bonferronis metod | Kapitel 7 |
| Tisdag 18/2 | Övning | Övningar i Kapitel 6 & 7: |
6.7.4, 6.7.5, 7.3.1, 7.3.5, 6.7.3, 7.3.4 (lösningar till kap. 6-7) |
| Onsdag 19/2 | Datorlab | For, If & Konfidensintervall | Lab 4 Rmd |
| Onsdag 19/2 |
Föreläsning (slides) |
Projekt & Försöksplanering | Kapitel 10a |
| Torsdag 20/2 |
Dugga |
Mid-term exam för bonuspoäng |
Questions with solutions |
| Tisdag 25/2 | Övning | Genomgång av mid-term |
|
| Onsdag 26/2 | Datorlab | t-test och ANOVA | Lab 5 Rmd Extra Fil |
| Onsdag 26/2 |
Föreläsning (slides) |
Enkel linjär regression, Hypotesprövning av lutning, Analys av residualer, Korrelation (Pearson), t-test av korrelation | Kapitel 8 |
| Torsdag 27/2 |
Föreläsning (slides) |
Frågestund, handledning av projekt | |
| Tisdag 4/3 | Övning | Övningar i Kapitel 8 / Handledning av projekt |
8.3.1, 8.3.2, 8.3.3, 8.3.4 (lösningar till kap. 8) |
| Onsdag 5/3 | Datorlab | Linjär regression | Lab 6 (Extra Fil) |
| Onsdag 5/3 |
Föreläsning (slides) |
Chi-två test, Chi-två fördelning, Goodness-of-fit test, Test av oberoende | Kapitel 9 |
| Torsdag 6/3 |
Föreläsning (slides) |
Försöksplanering och beräkning av stickprovsstorlek (& Projekt) | Kapitel 10b |
| Tisdag 11/3 | Övning | Övningar i Kapitel 9-10 |
9.3.1, 9.3.2, 9.3.4, 10.7.1 (lösningar till kap. 9-10) |
| Onsdag 12/3 | Datorlab | Chi-två-test | |
| Onsdag 12/3 |
Föreläsning (slides) |
Sammanfattning, genomgång av "vilket test?" | |
| Torsdag 13/3 |
Föreläsning (slides) |
Genomgång av gammal tenta, Frågestund, projekt | |
| Fredag 21/3 | Tentamen |
Course literature
We use the Kompendium written by Staffan Nilsson, Aila Särkkä, Serik Sagitov och Malin Palö Forsström. You can download it here. Answers to Kompendium exercises are here: Svarar. This Kompendium is under construction - notify Eszter Lakatos if you find any mistakes in it.
In addition to the kompendium, you can also read a basic book in statistics, e.g. Milton & Arnold: Introduction to probability and statistics.
Assessment
The assessment consists of two main parts, computational assignments and a written exam for the theoretical part (6 credits); and a project work and report (1.5 credits).
Computer lab assignments: you need to submit the solution to at least 3 of the extra exercises from computer labs, no later than 17 March. Each exercise counts separately, e.g. X5.1 and X5.2 are counted as 2. You can find all exercises in this Rmd file, and also at the end of each lab file. Submit your solutions in Canvas within the edited Rmd file. (Extra data files are here: X3.1, X5.3, X6.2, X7.2) You can submit the assignment through Canvas (in the Assignments tab), if you run into difficulties, you can
Mid-term exam: there is a mid-term exam on 20 February for bonus points. It follows the same format as the final exam. You can get 0-3 bonus points (up to 10% final exam points) in the mid-term.
Final exam: the exam consists of problems similar to those covered in exercise sessions. One paper (2 A4 pages) of notes can be used. Grades U, G or VG are given.
Project
The aim of the project is for you (in groups of 3-4 people) to try to plan an experiment which will then be analysed using the statistical methods you learned on the course. Research should be fun, so choose something that interests you! It is of course best if you can come up with a project within your main subject, but it is also fine with something that you just find interesting in general. You can choose to carry out experiments of some kind, or make an observational study, ask others to fill out a questionnaire, etc.
Minimum requirement for the project: you generate data, describe the data with suitable measures and plots and perform at least one statistical test. Significant results cannot be guaranteed, but you should at least have a research hypothesis that can reasonably be true.
Some examples of what has been studied in previous projects:
Use of Swish, Punctuality in trams, Tobacco consumption, Gingerbread test, Organic consumption, Price differences between certified and regular products, Eating habits (veg, fish, meat), Julmust test, Coffee consumption, Climate changes in Oslo since the 1930s, Social media and stress, Students' lunch habits, How Mariekex cracks (model for tectonic plates), Traffic flow and congestion tax, Sleeping habits, Dendochronology in Skogaby, etc.
Agree on the project idea with the course leader before starting the project.
Report: must contain a introduction describing the research question and the data, a method section where you describe the experimental setup, the method(s) you have used and give your null and research hypotheses, a results section where the results of your analyses are presented and a concluding discussion. The results can be described in plain text and with self-constructed tables, but also attach the R output that is the basis for the production, as well as the data files you used. In the discussion, you should discuss assumptions you have made and whether they are correct and limitations of your research.
You can find a minimal example of a report here.
The report should be emailed to the course leader no later than April 4, 2024. Only one person in the group needs to submit the report. Do not forget to enter the names of all group participants on the report.
Grades U or G are given for the project.
Use of AI tools
During the course work, feel free to use AI tools to support learning. However, be aware that AI generated code and text is usually too general and does not fully solve the problem at hand - always double check AI's answer and iterate using your own knowledge. During the exams it is forbidden to use any kind of AI tools.