Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.
Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.
Author ORCID Identifier
https://orcid.org/0009-0004-5226-2317
AccessType
Open Access Dissertation
Document Type
dissertation
Degree Name
Doctor of Philosophy (PhD)
Degree Program
Mathematics
Year Degree Awarded
2023
Month Degree Awarded
September
First Advisor
John Staudenmayer
Second Advisor
Anna Liu
Third Advisor
Ted Westling
Fourth Advisor
Chi-Hyun Lee
Subject Categories
Applied Statistics | Multivariate Analysis | Statistical Methodology | Statistical Models | Statistical Theory | Survival Analysis
Abstract
This dissertation is in the field of Nonparametric Derivative Estimation using
Penalized Splines. It is conducted in two parts. In the first part, we study the L2
convergence rates of estimating derivatives of mean regression functions using penalized splines. In 1982, Stone provided the optimal rates of convergence for estimating derivatives of mean regression functions using nonparametric methods. Using these rates, Zhou et. al. in their 2000 paper showed that the MSE of derivative estimators based on regression splines approach zero at the optimal rate of convergence. Also, in 2019, Xiao showed that, under some general conditions, penalized spline estimators of mean regression functions achieve optimal L2 rates of convergence. We extend this result to derivative estimators. In particular, we show that under similar conditions, penalized spline estimators of derivatives of mean regression functions achieve optimal L2 rates of convergence. In the second part of the thesis, we estimate the amount of association between
physical activity and all-cause mortality in US adults using penalized splines. We
introduce a novel nonparametric isotemporal substitution model to investigate the
dose-response relationship between daily time allocations across physical activity and
sedentary behaviors, and all-cause mortality. Our method reveals that the association between such daily time allocations and mortality depends on one’s level of physical activity. We apply our method to data from the 2003-2006 wave of the US National Health and Nutrition Examination Survey (NHANES) with mortality follow-up through December 31st, 2019, a nationally representative survey. Among US adults with less than 6 hours of daily activity, replacing 1 hour of physical activity with sedentary behaviors is associated with up to 68% increase in mortality risks after adjusting for sleep time and baseline demographic and health covariates. In addition, for those with sedentary time above 50% of non-sleep time, replacing 1% of moderate-to-vigorous activity (MVPA) time with sedentary time is associated with up to 18% increase in mortality risks. Therefore, to better understand mortality risk associations, US adults may consider their full daily activity time allocations before replacing one activity type with another.
DOI
https://doi.org/10.7275/35962201
Recommended Citation
Antwi Boasiako, Bright, "NONPARAMETRIC DERIVATIVE ESTIMATION USING PENALIZED SPLINES: THEORY AND APPLICATION" (2023). Doctoral Dissertations. 2947.
https://doi.org/10.7275/35962201
https://scholarworks.umass.edu/dissertations_2/2947
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Included in
Applied Statistics Commons, Multivariate Analysis Commons, Statistical Methodology Commons, Statistical Models Commons, Statistical Theory Commons, Survival Analysis Commons