Please note! Course description is confirmed for two academic years, which means that in general, e.g. Learning outcomes, assessment methods and key content stays unchanged. However, via course syllabus, it is possible to specify or change the course execution in each realization of the course, such as how the contact sessions are organized, assessment methods weighted or materials used.

LEARNING OUTCOMES

After passing the course the students can analyse and forecast time series using regression models and ARIMA-models. Students are able to apply linear regression model to analyse and forecast dependent variable under the model assumptions. In addition, the students are able to conduct diagnostic tests to validate the model assumptions. Students are familiar with the concept weakly stationary processes and they understand the most important related concepts including the autocorrelation function, partial autocorrelation function, and the spectral function. Students are also able to apply these functions in analysing real time series, for example, in recognising seasonal fluctuations. After taking the course, the students know ARIMA-models and their key properties. In addition, the students are able to model and predict the future behaviour of observed time series using ARIMA-models. The students also know basic concepts of dynamic regression models.

Credits: 5

Schedule: 26.10.2020 - 15.12.2020

Teacher in charge (valid 01.08.2020-31.07.2022): Pauliina Ilmonen

Teacher in charge (applies in this implementation): Pauliina Ilmonen

Contact information for the course (valid 22.09.2020-21.12.2112):

The lecturer of the course is Pauliina Ilmonen, pauliina.ilmonen@aalto.fi

The head assistant of the course is Nourhan Shafik, nourhan.shafik@aalto.fi

CEFR level (applies in this implementation):

Language of instruction and studies (valid 01.08.2020-31.07.2022):

Teaching language: English

Languages of study attainment: English

CONTENT, ASSESSMENT AND WORKLOAD

Content
  • Valid 01.08.2020-31.07.2022:

    The course is an introduction to time series analysis. Course topics include linear regression model and its diagnostics, central concepts of weakly stationary processes, ARIMA-models and their properties, stationarity of ARIMA-models, forecasting with ARIMA-models, Kalman filter, and introduction to dynamic regression models. Software R is used in the exercises of the course.

  • Applies in this implementation:

    Please note that all the lectures and exercises of this course are given in zoom. The lectures are not recorded, but the lecture slides are posted under "Materials"! Please note also that homework assignments are part of the grading. Please submit your solutions on time!


    After passing the course the students can analyse and forecast time series using regression models and ARIMA-models. Students are able to apply linear regression model to analyse and forecast dependent variable under the model assumptions. In addition, the students are able to conduct diagnostic tests to validate the model assumptions. Students are familiar with the concept weakly stationary processes and they understand the most important related concepts including the autocorrelation function, partial autocorrelation function, and the spectral function. Students are also able to apply these functions in analysign real time series, for example, in recognizing seasonal fluctuations. After taking the course, the students know ARIMA-models and their key properties. In addition, the students are able to model and predict the future behaviour of observed time series using ARIMA-models. The students also know basic concepts of dynamic regression models.

    Language of Instruction: English, All the lecture materials, exercises, exams etc are in English only!


Assessment Methods and Criteria
  • Valid 01.08.2020-31.07.2022:

    Homework assignments, exercise points, exam

  • Applies in this implementation:

    Evaluation: 1-5


    Exercises:

    Exercises are given in zoom. Exercises are an important part of this course. Attendance to zoom exercises is not compulsory, but highly recommended. You get exercise points by doing your homework assignments and submitting them on time. Zoom exercises might help in solving the problems and correct solutions to homework problems are provided in the online exercises only. The exercise points are valid until the end of June 2021.

    There are several exercise groups. Please attend one of the theory exercise groups and one of the computer exercise groups.

    How to pass this course?

    You are expected to

    Study the lecture slides carefully.

    Participate to exercises and solve your homework problems - not compulsory, but highly recommended - max 12 points.

    Take the exam - max 24 points.

    Max total points = 24 + 12 = 36: You need at least 18 points (or 12 points from the exam) in order to pass the course.




Workload
  • Valid 01.08.2020-31.07.2022:

    Lectures 24h (2), Exercises 24h (2), Homework assignments 48h, reading and studying the lecture materials 36h

DETAILS

Study Material
  • Valid 01.08.2020-31.07.2022:

    Lecture slides and the textbook Peter J. Brockwell, Richard A. Davis: Time Series Theory and Methods, Springer 2009 (reprint of the 2nd edition 1991).

  • Applies in this implementation:

    Please note that all the lectures and exercises of this course are given in zoom. The lectures are not recorded, but the lecture slides are posted under "Materials"! Please note also that homework assignments are part of the grading. Please submit your solutions on time!

Substitutes for Courses
  • Valid 01.08.2020-31.07.2022:

    Mat-2.3128 Prediction and Time Series Analysis

Prerequisites
  • Valid 01.08.2020-31.07.2022:

    MS-A05XX First course in probability and statistics, MS-A02XX Differential and integral calculus 2, and MS-A00XX Matrix algebra.