Time Series

Last updated on Dec 12, 2024

In the winter term we’ll offer the lecture “Time Series” (6 ECTS).

General

This course provides a comprehensive introduction to time series analysis, emphasizing both classical statistical techniques and modern machine learning methods. Students will learn to model, analyze, and forecast time series data using various approaches, equipping them with practical skills to tackle real-world time series problems.

Course Objectives:

  • Understand the fundamental concepts of time series analysis.
  • Apply classical statistical methods to time series data.
  • Utilize machine learning techniques for time series forecasting.
  • Implement and evaluate time series models.
  • Interpret the results and make informed decisions based on time series analysis.

Prerequisites:

  • Basic knowledge of statistics and probability.
  • Familiarity with linear algebra and calculus.
  • Proficiency in a programming language (preferably Python).

Organisation

Credits: 6 ECTS

Language: English

Lecture: Wednesdays, 8am - 10am c.t., Hörsaal F119, Sand

Tutorial: Friday, 8am - 10am (starting in the second week), Lecture Hall TTR2 (Maria-von-Linden Str 6, ground floor)

more information will be available on Moodle soon.

Topics to be covered (preliminary)

  1. Time Series Analysis (Smoothing, lag operators, stationarity, etc.)
  2. Linear Time Series Models (ARIMA, etc.)
  3. Non-Linear Time Series Models (GARCH, etc.)
  4. Multi-Variate Time Series Models (VAR, etc.)
  5. Forecasting, model evaluation and comparison
  6. Filtering and State Space Models
  7. Bayesian Approaches and Proper Scoring Rules
  8. Deep Learning Models
  9. Current Research
overview