Functional Data Workshop - Copenhagen, 22-26 September 2008

James Ramsay, McGill University

• Workshop outline
• Preparing for the workshop
• Audience
• Location
• Organizer
• Sign up
• Detailed program
• Resources (UPDATED)
• List of participants (UPDATE)

• What are functional data?
• How should they be prepared for analysis?
• How do we convert discrete noisy data to smooth functions?
• What data exploration tools are useful?
• Do the data display both phase and amplitude variation?
• What about principal components analysis and other exploratory methods?

The second two days will be primarily given over to linear models for functional data and to the modeling of dynamic systems. These are vast topics, and includes relatively basic topics like functional versions of analysis of variance and regression analysis and as well as issues less familiar to statisticians such as how differential equations can be used to model functional data.  All approaches assume that the goal is to explain variation in one or more response variables by variation in one or more input or independent variables where, naturally, at least one of the variables involved is functional. A detailed program is given below.

The last day of the workshop is dedicated to a small research workshop which aims at bringing people together to present some FDA related problems/topics for an open discussion. Please follow this link for details (and please notice that you shold sign up for this workshop separately).

Preparing for the workshop:
Please look at the material provided under Resources before the workshop. The practicals will be based on R. You are requested to bring your own laptop with the most recent version of R installed together with the R packages fda, spline, doBy and lattice.

Location:  
The first 4 days will take place at University of Copenhagen, Institute of Public Health, Østre Farimagsgade 5,  DK-1014,  Copenhagen  K (room 2.0.18).  The  last day will take place at University of Copenhagen, Faculty of Life Sciences, Thorvaldsensvej 40, Frederiksberg (Auditorium 3-14) 

Organizer:
The workshop is organized by the Danish Graduate School of Biostatistics. Course director is Søren Højsgaard (sorenh@agrsci.dk) 

Sign up:
Participation in the workshop is free and space is allotted on a first-come-first-serve basis. To sign up for the workshop, please send an email to Søren Højsgaard (sorenh@agrsci.dk) no later than June 20, 2008.

• Introduction to the workshop
• R-preliminaries
  
• An introduction to Functional Data Analysis. Slides Movies
• Studying human growth
• Canadian weather
• Non-durable goods
  
• From functional data to smooth functions. Slides Movies   
• Basis basics 
• Methods for representing functions
• Data smoothing with roughness penalties

• Spiegel-im-spiegel and FDA:  The dynamics of music appreciation
• Constrained optimization
• Curve registration
• Montreal weather:  Separating phase from amplitude variation
• Chinese handwriting
  

• Functional principal components analysis
• Canonical correlation analysis
• An overview of functional linear models
• Mouse livers: Modeling an input/output system

• Modeling dynamics:  Some examples from engineering and medicine
• A differential equation for the analysis and recognition of handwriting
• Analyzing lupus data:  fitting a differential equation to complex dynamics
• Resources:  web sites, books, data, and software
• Wrap-up: Themes and Challenges.

Resources:  

• Notes on Functional Data Analysis by James Ramsay.
• Copenhagen_Workshop.zip: A zip-file with various material.
• Chapters from book Functional Data Analysis with Matlab and R
• R in a few hours - Yet another short R introduction
  

• Ramsay, J. O. and Silverman, B. W. (1997) Functional Data Analysis. New York: Springer. 
• Ramsay, J. O. and Silverman, B. W. (2002) Applied Functional Data Analysis. New York: Springer.  (A book of case studies). 

• A short description by Bernard Silverman
• A functional data analysis web site: http://ego.psych.mcgill.ca/misc/fda/ 
• Introduction to MATLAB for Functional Data Analysis

Participants: