Machine Learning for Time-Series Data

Catalog description

This course is designed to equip students with tools for analyzing and forecasting time-series data. The course starts with the fundamental principles of time-series analysis, including classical models like ARIMA, state-space models, and frequency domain analysis (Fourier and Wavelets). Students will then delve into machine learning methods, starting with random forests and gradient boosting for time- series, following with learning techniques such as RNNs, LSTMs, and GRUs. Recent techniques, such as attention mechanisms and transformers will be covered at the end of the course. The course emphasizes both theoretical understanding and practical application, offering hands-on experience with real-world datasets and tools like Python and TensorFlow. This curriculum is tailored for students with a background in engineering, computer science, or finance, preparing them for advanced research and professional roles in time-series analysis.