Searching for Patterns by Prof. Heggeseth

Friday, November 15, 2013 at 1:00pm to 1:45pm

Bronfman Science Center, 106 18 Hoxsey St, Williamstown, MA 01267, USA

Searching for Patterns by Prof. Heggeseth

Searching for Development Patterns:  An Application to Childhood Growth Trajectories by Prof. Brianna Heggeseth

Mathematics and Statistics Department Faculty Seminar

Abstract:  A key advantage of longitudinal data is the ability to study how individuals change over time. My work revolves around discovering distinct development patterns by clustering or grouping similar individuals. The majority of available cluster analysis methods for this type of structured data do not group individuals based on the most interesting characteristic of the data: the shape of the trajectory over time.  I will discuss two standard clustering approaches used for longitudinal data and then introduce three methods that define similarity with respect to the development patterns. Throughout the talk, these methods will be compared using a simulated data set as well as growth patterns of children enrolled in the Center for the Health Assessment of Mothers and Children of Salinas (CHAMACOS) study.

Event Type



Mathematics & Statistics

Recent Activity

People Going

Getting Here