Date of Award
Master of Science (MS)
College of Science and Mathematics
Thesis Sponsor/Dissertation Chair/Project Chair
Andrada E. Ivanescu
Helen M. Roberts
The prediction of functional data samples has been the focus of several functional data analysis endeavors. This work describes the use of dynamic function-on-function regression for dynamic prediction of the future trajectory as well as the construction of dynamic prediction intervals for functional data. The overall goals of this thesis are to assess the efficacy of Dynamic Penalized Function-on-Function Regression (DPFFR) and to compare DPFFR prediction intervals with those of other dynamic prediction methods. To make these comparisons, metrics are used that measure prediction error, prediction interval width, and prediction interval coverage. Simulations and applications to financial stock data from Microsoft and IBM illustrate the usefulness of the dynamic functional prediction methods. The analysis reveals that DPFFR prediction intervals perform well when compared to those of other dynamic prediction methods in terms of the metrics considered in this paper.
Rios, Nicholas, "Prediction Intervals for Functional Data" (2018). Theses, Dissertations and Culminating Projects. 1.