This course introduces methods in functional data analysis, with an emphasis on practical issues and applications.

15 lectures (1 or 2 each week)

5 2-hour computer-based practicals

STATS4073 Functional Data Analysis

Assessment

Project work (25%) and final examination (75%)

To introduce the students to functional data analysis methods applied to a wide array of application areas;

To illustrate common numerical and estimation routines to perform functional data analysis;

To demonstrate applications where functional data analysis techniques have clear advantage over classical multivariate techniques.

By the end of this course students will be able to:

■ identify scenarios where data may be considered to be smooth functions and apply functional data analysis techniques;

■ construct visualization strategies and implement nonparametric smoothing for exploring functional data;

■ formulate and fit several types of functional linear models;

■ describe the functional principal component analysis algorithm and apply it in simple cases;

■ construct suitable methods for analysis involving derivatives and apply these techniques to provide solutions to practical problems;

■ discuss the principles behind registration and apply this technique to practical problems where registration is a crucial pre-processing step;

■ Understand one of the above topics in depth.