12.864 Inference from Data and Models, Spring 2004
First solution basis vector obtained in solving the Laplace equation using the singular value decomposition. This field arises from the need to solve practical problems with incomplete, contradictory and erroneous data and is an example of an inverse method. (Image courtesy of Prof. Carl Wunsch.)
Highlights of this Course
This course features, in the lecture notes section, a primer that complements the course textbook, along with references and exercises. Both the textbook (The Ocean Circulation Inverse Problem) and primer on Time Series Analysis have been authored by the instructor. Homework assignments are also available.
Course Description
The course is directed at making scientifically sensible deductions from the combination of observations with dynamics and kinematics represented, generically, as "models". There are two overlapping central themes
-
Linear "inverse" methods and data "assimilation" including regression, singular value decomposition, objective mapping, non-stationary models and data, Kalman filters, adjoint methods ("assimilation") etc.
-
Standard time series analysis, including basic statistics, Fourier methods, spectra, coherence, filtering, etc.