lu.se

Matematikcentrum

Lunds universitet

Denna sida på svenska This page in English

Time series analysis (FMSN45/MASM17)

News

  • Clarification: if you solve part B, and get 5 marks, that will count as pass even if you do not solve part A. Also, in case one fails the exam, there will be a possibility to complement the exam. Although, in that case, one can only get a pass grade on the course. Sorry for not being clearer about this.
  • Found typo: for some reason I managed to put twice the number of marks on the exam than there should be... Sorry about the confusion. The total should be 30 marks, as stated before, not 60... The exam is now corrected.
  • Found typo: in problem 3b in the take-home, it should read x(t-2), not x(t-1), to make any sense. The pdf is now updated.
  • The take home exam is now available (see link below).
  • Andreas will have an extra office hour slot on January 9th, at 2-3 pm.
  • Per Niklas will have an extra office hour slot on Thursday, December 19th, at 3-4 pm.
  • Filip has written two notes to help explaining the transfer function model. You can find them here and here.
  • You can now sign up for lab 2 and 3 (see link below).
  • The project is now available (see link below).
  • You can now sign up for the first lab (see link below).
  • The course book (3rd edition) is available at KFS. It can also be ordered online. You will need it.

Course contents

Time series analysis concerns the mathematical modeling of time varying phenomena, e.g., ocean waves, water levels in lakes and rivers, demand for electrical power, radar signals, muscular reactions, ECG-signals, or option prices at the stock market. The structure of the model is chosen both with regard to the physical knowledge of the process, as well as using observed data. Central problems are the properties of different models and their prediction ability, estimation of the model parameters, and the model's ability to accurately describe the data. Consideration must be given to both the need for fast calculations and to the presence of measurement errors. The course gives a comprehensive presentation of stochastic models and methods in time series analysis. Time series problems appear in many subjects and knowledge from the course is used in, e.g., automatic control, signal processing, and econometrics.

Higher education credits: 7,5 Level: A

Language of instruction: The course will be offered in English if non-Swedish speaking students are attending.

Prerequisites: Basic courses in probability and statistics, as well as stationary stochastic processes.

Literature: Andreas Jakobsson, An Introduction to Time Series Modeling (3rd edition), Studentlitteratur, 2019.

Time: Lectures are held Mondays and Wednesdays 13-15. Exercises are held Thursdays and Fridays; please see the detailed schedule.

Office hours: Andreas will have office hours in MH:217 on Wednesdays 10-12 (until 11/12). There will also be additional office hours on Thursday 12/12 at 9-12. FIlip will have office hours in MH:138 on Tuesdays 9-10 (until 4/12). Per Niklas will have office hours in MH:223 at 10-11 on 5/12 and 10/12, as well as 10-12 on 16/12. Without appointment, please respect these hours.

Note: Both Filip and Andreas will be unavailable (attending a conference) during December 13-19. Please plan accordingly!

Course material

General material:

Lecture notes and schedule:

  • Week 1:
    • L1: Introduction and overview. Multivariate random variables. [slides 1, 2]
    • L2: Multivariate random variables. Stochastic processes. [slides 3, 4, 5]
    • Reading instructions: Ch. 1, 2, 3.1-3.3
    • Textbook problems: 2.1-2.3, 3.1-3.4
    • Mini project: [pdf, data]
  • Week 2
    • L3: Stochastic processes. [slides 1, 2, 3]
    • L4: Stochastic processes. Identification. [slides 4, 5]
    • Reading instructions: Ch. 3, 4.1-4.2
    • Textbook problems: 3.5-3.10, 3.12-3.15
    • Mini project: [pdf, data]
  • Week 3
    • L5: Identification. [slides 1, 2, 3]
    • L6: Estimation. [slides 4, 5]
    • Reading instructions: Ch. 4, 5.1-5.2
    • Textbook problems: 4.1-4.4
    • Mini project: [pdf, data]
  • Week 4
    • L7: Estimation. Model order selection. [slides 1, 2]
    • L8: Residual analysis.
    • Reading instructions: Ch. 5
    • Textbook problems: 5.1-5.5, 5.8, 5.10-5.11
    • Computer exercise 1 (see below).
  • Week 5
    • L9: Prediction. Multivariate time series. [slides 1, 2]
    • L10: Multivariate time series. [slides 3]
    • Reading instructions: Ch. 6, 7
    • Textbook problems: 6.1-6.8
    • Computer exercise 2 (see below).
  • Week 6
    • L11: Recursive estimation. State space models. [slides 1]
    • L12: The Kalman filter. Project discussion.
    • Reading instructions: Ch. 8
    • Textbook problems: 7.1-7.4, 8.1-8.2
    • Computer exercise 3 (see below).
  • Week 7
    • Textbook problems: 8.3-8.8

Examination

The course examination consist of mandatory computer exercises, a take-home exam, as well as a project. As a part of the examination, a detailed project report should be handed in, as well as the result being disseminated in an oral presentation (about 10 minutes long).

Further details:

  • Computer exercises:
    • You need to sign up for the computer exercises - you can do this here. If you have not signed up for an exercise, you may only attend the session if there are available slots left; otherwise, you will be asked to leave.  
    • Exercise 0. [Do this on your own]
    • Exercise 1.
    • Exercise 2.
    • Exercise 3.
    • Please be aware that you are expected to come well prepared to the computer exercises. If you have not, you may be asked to leave.
  • Project:
    • The project is available here. The data for the project is available here.
    • The project examination will take place on 20/12, at 13-16, in MH:R, or on 17/1, at 13-16, in MH:R. Choose either of these times; you cannot attend without being ready to present.
    • The project report and the presentation material should be handed in no later than at the start of the presentation. Printed versions of the project report and the take home should be handed in to the course secretary. The slides for the presentation may be mailed as a pdf to the lecturer directly.
  • Take home exam:
    • The take-home exam is available here. The exam is due on 21/1, at 13.15.

Advanced courses

After completing this course, you may be interested in the following courses:

Course start

First Lecture:
Monday, November 4, 2019, at 13.15, in M:B

Reading periods:
HT2

Contact

Lecturer: 

Andreas Jakobsson, 046-222 4520

Teaching assistants: 

Filip Elvander, Per Niklas Waaler, Amanda Nilsson, Wilhelm Ålander, Erik Wik

Course secretary: 

Susann Nordqvist, 046-222 4577