Functional Data Analysis in a Nutshell
Functional data analysis (FDA) is a field of statistics that deals with the analysis of data that have a functional character. Functional data include curves, images, surfaces and trajectories. In the following, we will focus on curves. Growth curves are an example for one-dimensional functional data observed over time. Other examples are spectrometric measures over wavelength or blood markers measured continuously over time. FDA is applied in diverse fields including biometry, demography, medicine, linguistics and finance. Instead of analyzing single points on the curves, FDA treats the curves as observation units. The talk will approach FDA rather intuitively to give an idea of functional data. The talk covers basic summary statistics, like mean and variance for functional data, and contains an outlook to more complex methods like regression with functional data.
Introduction to FDA:
Ramsay & Silverman (2005), Functional data analysis, Springer, New York.
Homepage of the book: http://www.psych.mcgill.ca/misc/fda/