Academics

  /  

Courses

  /  

Descriptions

ES_APPM 479: Data Driven Methods for Dynamical Systems

Description

The course will survey methods for characterizing time-series data by reading primary literature and implementing and testing methods on synthetic data. Students will simulate time-series from a variety of deterministic and stochastic systems and evaluate a range of methods. A goal of the course is to understand the suitability of different methods for characterizing systems with noise, nonlinearities, and other dynamic characteristics. Topics will include Granger causality, Taken's embedding theorem, convergent cross mapping, entropy-based methods, sparse model selection, dynamic mode decomposition, Koopman operators. 

No required textbook.