Speaker: James Matuk, Ohio State University (Wed. 03/13)
- 2:00 PM to 3:00 PM
- COLH 446
Functional and Shape Data Analysis
Functional and shape data analysis is a field in statistics that concerns itself with data that takes the form of functions from a time domain D to R^k. This format encompasses data sources that are used in many scientific fields including biomarkers in medicine, functional magnetic resonance imaging time courses in neuroscience, and signature data and image contours in computer vision. While these data are rich with information, they are difficult to work with because of their distinctive nature.
In this talk we will discuss some important problems in this research area including the main types of variability that these data are subject to, the visualization of central tendency and dispersion, and considerations for formulating statistical models for inference.