Home Page for course OR779
School of Operations Research and Industrial Engineering
Cornell Unversity

Functional Data Analysis

Links to Lectures (with summary of topics):

Student Presentation 5/1/02:    Min Zhang

Student Presentation 4/29/02:    Xin Zhao

Student Presentation 4/24/02:   Jing Qiu

Lecture 4/22/02    (Finished ICA for discrimination, Polynomial Embedding, Kernel Embedding, Support Vector Machines, Validation of Discrimination)

4/17/02   Class canceled

Student Presentation 4/15/02:   Rommel Regis

Lecture 4/10/02    (High dimensional space is strange, FLD expanding dimensions, FLD within class vs. global covariance estimates, ICA for discrimination, toy examples, Corpora Collosa data)

Lecture 4/8/02    (Introduction to HDLSS statistics, Review of FLD vs. Mean Difference, New conceptual model for HDLSS methods, nature of high dimensional Gaussian data)

Lecture 4/3/02    (Fisher Linear Discrimination: Mahalanobis distance, likelihood view.  Generalizations: general Gaussian likelihood ratio, multi-class & Principal Discriminant Analysis, FLD for Corpora Collosa data)

Lecture 4/1/02    (Finished ICA nonlinearity toy examples, Big Picture View of Course Material, Discrimination (i.e. classification), simple methods and Fisher Linear Discrimination, "sphering transformation" derivation)

Student Presentation 3/27/02   Trevor Park,  Varimax rotation of PCA

Lecture 3/25/02    (Independent Component Analysis, toy examples, Curve Data examples & contrast with PCA, numerical issues and choice of "nonlinearity")

Student Presentation 3/13/02   LongYu,  PCA and Smoothing

Lecture 3/11/02    (Independent Component Analysis, algorithm, non-Gaussianity, Q-Q plots, toy examples)

Lecture 3/06/02    (finishedSIZER background, started Independent Component Analysis)

Lecture 2/27/02    (SIZER background, which features are "really there"?)

Student Presentation 2/25/02   Hui-Bin Zhou,  Shirnkage Estimation

Lecture 2/20/02    (PCA for boundary Fourier Corpora Collosa, intro and PCA for M-Rep Corpora Collosa, correlation PCA, PCA and clusters, mass flux data)

Lecture 2/18/02    (Dual Eigen Decompositions, statistics of PCA, intro to Corpora Collosa data with boundary Fourier representation)

Lecture 2/13/02    (Review of Linear Algebra, Singular Value and Eigen Decompositions, Multivariate Probability)

Lecture 2/11/02    (Elliptical Principal Component Analysis, Cornea Data, Another Toy Example, Start Deeper look at PCA, Review of Linear Algebra)

Lecture 2/6/02    (Robust, Spherical and Elliptical Principal Component Analysis)

Lecture 2/4/02    (Principal Component Analysis of Cornea Data, Introduction to Outliers and Robust Statistics)

Lecture 1/30/02    (Principal Component Analysis for curves, toy 3-d and 10-d examples, including Dog Legs, Fans, Parabolas, Gaussians and 2 clusters)

Lecture 1/28/02    (Introduction to Principal Component Analysis, careful 2-d example, intuition from curve view)

Lecture 1/21/02    (Introduction to Functional Data Analysis, Data Representation, Object Space - Feature Space duality, Main goals: "understanding population structure" & discrimination (i.e. classification))

Link to Combined References



Course Meetings:

     Time:   Mon. - Wed. 8:40 - 9:55
     Room:  Rhodes 471

Course Web Site:


maybe easier to follow link from:


Instructor:   J. S. (Steve) Marron

Office:   Rhodes 234
Office Hours:   Mon. 10 - 11,    Tuesday 11 - 12

Phone:   (607) 255-9147
Email:   marron@stat.unc.edu

Course Email List: please add yourself,

Subscribe to or778-sp02-l, by sending a message with the word subscribe in the Subject: field to the address:     or778-sp02-l-request@orie.cornell.edu

To unsubscribe, send a message with the word unsubscribe in the Subject: field to or778-sp02-l-request@orie.cornell.edu

                    (useful for announcements, such as "notes now posted")

Recommended Textbook:

Ramsay, J. O. & Silverman, B. W. (1997) Functional Data Analysis, Springer, N.Y.

Course Work / Grading

Based on a presentation

Presentations:  can be any of (you choose, or I suggest):

    -    a section of Ramsay and Silverman

    -    a paper by others

    -    your own work

Let's discuss soon