This hands-on workshop introduces participants to the core methods and modern techniques in time series analysis. Spanning from classical decomposition and forecasting to machine learning and deep learning models, the workshop integrates theory with practical coding exercises in Python.
Learning Objectives
- Understand and decompose time series into basic components
- Test stationarity and transform non-stationary data
- Build classical models like: ARIMA, SARIMA
- Explore multivariate dependencies using VAR and ARIMAX
- Apply classic machine learning models for forecasting and anomaly detection
- Introduction to deep learning architectures like RNNs, LSTMs, and GRUs for time series predictions
Prerequisites
Participants should have:
Basic knowledge of statistics (mean, variance, correlation)
Prior experience with Python (pandas, numpy, matplotlib)
Target Audience
Graduate students, researchers, and professionals in any discipline using time-related data, seeking to enhance their analytical and forecasting capabilities
Dates & Place
Sunday 19.10.2025 & Wednesday 22.10.2025, 09:00-16:00
Registration Cost
700 NIS - Regular price for faculty member at another university, industry or others.
400 NIS - Reduced price for students at another university.
The workshop is free for students or faculty members of the University of Haifa
