MVE550 Stochastic processes and Bayesian inference Autumn 24
Course PM
This page contains the program of the course. Other information, such as learning outcomes, teachers, literature, examination, and old exams, are in a separate course PM.
Program
The schedule of the course is in TimeEdit and in the table below. Each week there are (generally) three lectures. There is also an exercise session on Thursdays. Starting the second week, Petter Mostad will have office hours on Tuesdays after the lecture.
The lectures will be possible to follow either in the lecture room or live on-line, and will be recorded. Recordings will be available from this Canvas page.
Details about the literature is found in the course PM; the relevant chapters for the different lectures are listed in the table below. Overheads and other material used in the lectures will be posted on Canvas (usually) before the lectures.
There will be two parallel exercise sessions, and you can choose to attend either (or none). The teachers will demonstrate solutions to specific exercises, listed in the table below, and there will be time for discussions and individual help. Lists of additional recommended exercises are also found below. Generally exercise sessions will focus on material from the lectures roughly one week earlier, so you have time between the lecture and the exercise class to work on the exercises.
The following sections of Dobrow are NOT part of the curriculum: 3.9, 3.10, 5.5, 5.6, 6.8, 9.1,9.2,9.3.
It will be possible to pose and discuss questions via Piazza, but to do so, you first need to enroll using this link.
Assignments
An important part of the course are the three obligatory assignments. They should be solved and submitted in small groups of 2-3 students. When answering these assignments it is of course not allowed to use AI tools. The groups are created by the students themselves. It may be a good idea to find a group directly, as you can then get started working together with somebody. One way to find group partners is to use "Discussions" here in Canvas. If you need help, you may contact Petter. Note that deadlines are at 16:00 on the relevant dates.
When your group is ready to submit answers to the first assignment, one group member goes to the "People" menu and the "Assignment groups" menu. Select a group without members, and then invite your group partners into the group. The answers are then submitted in the name of this group. This ensures that all group members get graded, and get feedback, as they should. Normally the same people will work together for all three assignments, but occasionally students switch groups, or drop out of the course. If this happens, you must mail me with complete details about who is working on the assignments, and who is not. Submitting answers in the name of a group that contains members that have not contributed is otherwise a kind of cheating. Unfortunately, Canvas does not allow changes to groups that have already submitted answers.
For more details on the assignments go to the list at the bottom of this page.
Some recommended exercises
Dobrow:
Chapter 1: 1, 3, 7, 17, 26, 33
Chapter 2: 1, 3, 5, 9, 11, 25
Chapter 3: 1, 3, 7, 11, 13, 19, 33
Chapter 4: 1, 3, 7, 17, 23, 33
Chapter 5: 1, 3, 7, 15
Chapter 6: 1, 3, 7, 9, 15, 19
Chapter 7: 3, 7, 11, 13, 21, 29
Chapter 8: 3, 5, 7, 11, 13, 19
NOTE: It is highly recommended to practice solving old exams: They are available, together with solutions, from the PM page.
In the compendium, all exercises are recommended, including those not listed in the schedule below. Solutions can be found in an Appendix of the compendium.
IMPORTANT: The selection of exercises above is intended as a study help. It is NOT the case that theory, techniques, or ideas mentioned in the lectures but not "trained" in the exercises above cannot appear on the exam! You are expected to know and understand theory well enough to be able to apply it even when you have not done exercises with it.
If you have questions about the exercises above (or any other exercises) you are welcome to ask questions at the exercise classes. In these classes, we will mainly go through together the separate list of exercises listed in the schedule below.
Lectures and exercise classes
| Contents | Study material | Meeting |
|---|---|---|
| Lecture 1: Introduction and review |
Dobrow Chapter 1 + Appendices A, B, C, D. Overheads. Recording. |
Tuesday 5/11 13:15 - 15:00, HB3. Zoom meeting. Passcode 691373 |
| Lecture 2: Basics of Bayesian inference |
Thursday 7/11 13:15 - 15:00, HC4. Zoom meeting. Passcode 691373 |
|
| Exercises |
Dobrow Chapter 1: 6, 9, 10, 14, 16, 20, 24, 30, 32, 35. |
Thursday 7/11 15:15 - 17:00, SB-L200 and SB-L216. |
| Lecture 3: Markov chains |
Friday 8/11 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
|
| Lecture 4: Markov chains |
Tuesday 12/11 13:15 - 15:00, HB3. Zoom meeting. Passcode 691373 |
|
| Office hours |
Ask questions to Petter Mostad |
Tuesday 12/11 15:15 - 17:00 in MVH3017 |
| Lecture 5: Markov chains |
Thursday 14/11 13:15 - 15:00, HB3. Zoom meeting. Passcode 691373 |
|
| Exercises |
Compendium Chapter 1: 3, 4, 6, 7. Dobrow Chapter 2: 2, 7, 10, 14, 16, 17, 18 |
Thursday 14/11 15:15 - 17:00. ML11 and ML13. |
| Lecture 6: Branching processes |
Friday 15/11 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
|
| Lecture 7: MCMC |
Thursday 21/11 13:15 - 15:00, HC4. Zoom meeting. Passcode 691373 |
|
| Exercises |
Dobrow Chapter 2: 20, 21, 22, 26. Chapter 3: 2, 5, 9, 14, 18, 22, 30, 35, 38, 41, 46, 64. |
Thursday 21/11 15:15 - 17:00. ML11 and ML13 |
| Office hours |
Ask questions to Petter Mostad |
Friday 22/11 10:00 - 11:45 in MVH3017 |
|
Lecture 8: MCMC for Bayesian inference |
Dobrow Chapter 5. Compendium Chapter 2. Overheads. Recording. R code. |
Friday 22/11 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
| Lecture 9: MCMC |
Lecture on MCMC and Bayesian inference. Overheads. Unfortunately no recording exists. |
Tuesday 26/11 13:15 - 15:00, HB3. Zoom meeting. Passcode 691373 |
| Lecture 10: Poisson processes |
Thursday 28/11 13:15 - 15:00, HC4. Zoom meeting. Passcode 691373 |
|
| Exercises |
Dobrow Chapter 4: 2, 5, 6, 8, 12, 19, 20, 26, 29 |
Thursday 28/11 15:15 - 17:00. ML11 and ML13 |
| Office hours |
Ask questions to Petter Mostad |
Friday 29/11 10:00 - 11:45 in MVH3017 |
| Lecture 11: Poisson processes | Dobrow Chapter 6. Overheads. Recording. |
Friday 29/11 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
| Lecture 12: Continuous-time Markov chains |
Tuesday 3/12 13:15 - 15:00, HB3. Zoom meeting. Passcode 691373 |
|
| Lecture 13: Bayesian parameter inference for stochastic processes |
Thursday 5/12 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
|
| Exercises |
Dobrow Chapter 5: 2, 4, 5, 6, 8, 9, 18. Chapter 6: 2, 5, 8 |
Thursday 5/12 15:15 - 17:00. ML11 and ML13 |
| Office hours |
Ask questions to Petter Mostad |
Friday 6/12 10:00 - 11:45 in MVH3017 |
| Lecture 14: Continuous-time Markov chains |
Friday 6/12 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
|
| Lecture 15: Continuous-time Markov chains |
Tuesday 10/12 13:15 - 15:00. HB3. Zoom meeting. Passcode 691373 |
|
| Lecture 16: Brownian motion |
Thursday 12/12 13:15 - 15:00. KE. Zoom meeting. Passcode 691373 |
|
| Exercises |
Compendium Chapter 3: 1, 2, 3, 4, 6. Dobrow Chapter 6: 10, 14, 16, 21, 24. Chapter 7: 1, 4, 5, 6, 12. |
Thursday 12/12 15:15 - 17:00. ML11 and ML13 |
| Office hours |
Ask questions to Petter Mostad |
Friday 13/12 10:00 - 11:45 in MVH3017 |
| Lecture 17: Brownian motion | Dobrow Chapter 8. Overheads. Recording. R code. | Tuesday 17/12 13:15 - 15:00. HB3. Zoom meeting. Passcode 691373 |
| Lecture 18: Brownian motion |
Thursday 19/12 13:15 - 15:00. HB3. Zoom meeting. Passcode 691373 |
|
| Exercises |
Dobrow Chapter 7: , 14, 18, 20, 23, 26, 30, 34. Chapter 8: 6, 9, 16, 18, 19, 22, 25, 26, 29 |
Thursday 19/12 15:15 - 17:00. SB-L200 and SB-L216 |
| Office hours |
Ask questions to Petter Mostad |
Friday 20/12 10:00 - 11:45 in MVH3017 |
| Review |
Friday 20/12 13:15 - 15:00. HC4. Zoom meeting. Passcode 691373 |
|
| Exam |
|
Monday 13/1 8:30-12:30 |
Course summary:
| Date | Details | Due |
|---|---|---|