Analysis of Object Data II

Principal Instructors: J. S. Marron, J. O. Ramsay

Course Day and Time: Course will be hold at SAMSI (driving directions) in RTP on Wednesdays, 4:30-7:00 p.m. in Room 150.

Schedule: First class Wednesday, January 19; last class Wednesday, April 27

Summary: This second course associated with the SAMSI program on Analysis of Object Data will cover topics that generally differ from those covered in the course taught in the Fall, and thus will not have that as a prerequisite. Following a brief overview of the basics of Functional and Object Data, three main segments of the course are planned.

1. Dynamic systems: (Ramsay)
Models for multivariate functional data that explicitly model change through the involvement of one or more derivatives in the model specification. Topics include the anatomy of a dynamic system, dynamic systems as extensions of functional models, and parameter estimation and inference for systems where analytical solutions are impossible.

2. Manifold data: (Ramsay and Marron)
This topic extends functional data analysis in a number of ways, where data are typically distributed over time and/or space, to situations where the data are distributed over manifolds embedded within higher dimensional spaces. Along with a quick review of classic subjects such as principal components analysis and test theory, more advanced topics include medial shape representations, diffeomorphisms in image analysis, and diffusion tensor imaging.

3. Tree-structured data: (Marron)
This section will contrast the very diverse combinatorial, folded Euclidean and Harris correspondence approaches to this new area of data analysis.

Registration for this course is being processed through your university.

Course Numbers:
Duke: 294-01
NCSU: MA 810.002
UNC: STOR 891-001

Questions about the course or the AOOD program should be emailed to aood@samsi.info.