Restricted Research - Award List, Note/Discussion Page
Fiscal Year: 2023
1835 The University of Texas at El Paso (143723)
Principal Investigator: Pokojovy,Michael
Total Amount of Contract, Award, or Gift (Annual before 2011): $ 386,308
Exceeds $250,000 (Is it flagged?): Yes
Start and End Dates: 7/1/22 - 8/31/25
Restricted Research: NO
Academic Discipline: Mathematical Sciences
Department, Center, School, or Institute: Mathematical Sciences
Title of Contract, Award, or Gift: Nonparametric Total Variation Regression for Multivariate Process Data
Name of Granting or Contracting Agency/Entity:
NATIONAL SCIENCE FOUNDATION
CFDA Link: NSF
47.049
Program Title:
Mathematical and Physical Sciences
CFDA Linked: Mathematical and Physical Sciences
Note:
We propose new nonparametric estimators, the total variation (TV) and the taut string (TS) estimators, to estimate the mean of a multivariate process and put forth a new class of multivariate statistical control charts for monitoring individuals process data in Phase II Statistical Process Control (SPC). The data are assumed to be sequentially collected from an independent sub-Gaussian process with a possibly varying mean vector in an abrupt or continuous fashion. Individuals multivariate control charts are commonly recognized as an indispensable statistical quality assurance tool used to monitor process data occurring in engineering, manufacturing, commerce, environmental studies and other arenas. The overarching goal of this project is to develop nonparametric statistical and algorithmic tools aimed at nonparametrically estimating the process mean identifying the presence and/or type of shift in the mean. Development of estimation and inference theories for these nonparametric methodologies is pursued. The proposed control charts will be capable of effectively monitoring multivariate individuals process data (i.e., single response at predetermined time points) for both gradual and abrupt changes in the mean. The project involves fundamental research in theory in nonparametric statistics, convex optimization and statistical process control.
Discussion:
Withdrawn by Institution (Manuela Dokie) - Previously reported in 2022.