BACKGROUND
Time series analysis is a critical skill in many scientific fields, including climate research, economics, health sciences, and more. Understanding and predicting time series data is essential for making informed decisions and uncovering hidden patterns in data.
Python has become a powerful tool in data science due to its extensive libraries and community support. Importantly, programming languages like Python offer the ability to seamlessly integrate and manage the entire data cycle, from data processing, analysis, and modeling to visualization and communication. Jupyter notebooks play a key role by promoting reproducibility and transparency, allowing scientists to document and share their analysis and visualization workflows.
WORKSHOP GOAL
Participants will learn the fundamentals of time series analysis and prediction using Python. The course combines engaging presentations on statistical theory with interactive, hands-on sessions where participants apply these concepts using Python. Participants will work through example problems focusing on the Earth Sciences and use popular Python libraries to process and analyze time series data.
By the end of the course, participants shall have the background statistical knowledge and programming skills to analyze and predict time series data effectively using Python and document their work using Jupyter.
WORKSHOP CONTENT
Day 1: Python crashcourse; Statistics and Python for time series analysis
- Introduction to time series analysis and fundamental concepts
- Quickstart with Python using Jupyter notebooks
- Python libraries for time series data processing and analysis
Day 2: Statistics and Python for time series prediction
- Introduction to time series prediction and fundamental concepts
- Models for stationary time series and trends
- Machine learning methods for time series prediction
Day 3: Practical exercise
TARGET AUDIENCE & PRIOR KNOWLEDGE
This course is suitable for researchers with all levels of Python knowledge and statistical background, although basic knowledge in at least one of these areas is an advantage. Researchers from all disciplines are welcome, though researchers from the Earth Sciences will benefit most from data use cases. The workshop contains hands-on sessions and thus limited to max. 20 participants.
TECHNICAL REQUIREMENTS
Connection to the Wifi (e.g. via eduroam -> https://www.uni-bremen.de/en/zfn/wifi/overview-wifi).
Own laptop with Python installed. Participants will receive installation instructions prior to the workshop.
ABOUT THE TRAINER
Dr. Maryam Movahedifar and Annika Nolte are data scientists for training and consulting at the DSC.
Maryam holds a PhD in Statistics and has extensive experience in Interpretable Machine Learning and Time Series Analysis. With a strong foundation in statistical methods and practical experience in applying these techniques to real-world problems, she is well-equipped to teach complex machine learning and time series concepts. Her expertise includes making advanced models understandable and accessible, ensuring that even the most complex analyses can be effectively communicated.
Annika holds a master’s degree in Environmental Sciences from the Technical University of Braunschweig (2019) and is a PhD candidate at the Universität Hamburg. Annika has worked with Python for more than 4 years on database development and analysis of quantitative environmental data. Her skills lie in scientific programming, hydroinformatics, GIS and geospatial analysis and AI in environmental research.