Date of Award

5-2019

Document Type

Thesis

Degree Name

Master of Science (MS)

College/School

College of Science and Mathematics

Department/Program

Mathematical Sciences

Thesis Sponsor/Dissertation Chair/Project Chair

Andrada E. Ivanescu

Committee Member

Andrew McDougall

Committee Member

Haiyan Su

Abstract

We study the data setting consisting of functional data sets repeatedly observed over time. The focus is on the dynamic prediction of the future trajectory for a subject. Regression methods based on dynamic functional models are used for dynamic prediction of individual trajectories. We propose strategies for the selection of the study sampling design in the context of longitudinal functional data. An application to simulated child growth data is presented. The height-for-age z-score (HAZ) was the response variable in the functional dynamic models for prediction. The intent was to recommend four months for removal in our initial historic data set. We quantify the effect on dynamic prediction performance when several data missing scenarios and methods of data imputation were considered. The effectiveness of seven methods of data imputation in the setting of longitudinal functional data were examined.

File Format

PDF

Share

COinS