February 5, 2026 Webinar
Multiple-use prediction and calibration for all future values: simultaneous or pointwise tolerance bands, or something else?
Yang Han, PhD
Abstract:
Multiple-use prediction and calibration for an infinite sequence of future observations play a valuable role in a wide range of fields, including health and medical research. Simultaneous tolerance bands (STBs) and simultaneous tolerance regions (STRs) can be used for this purpose. Motivated by a real-world example involving the estimation of gestational age using ultrasound measurements, this study focuses on the construction of exact STBs for polynomial regression and STRs for multivariate regression over any given covariate interval. Furthermore, we introduce weighted simultaneous tolerance regions (WSTRs), and demonstrate that confidence sets based on WSTRs exactly satisfy the key property required for multiple-use calibration. Through the lens of WSTRs, we show that confidence sets based on pointwise tolerance regions do not guarantee this property.
Short Bio:
Dr. Yang Han is a Senior Lecturer (Associate Professor) in Statistics in the Department of Mathematics at the University of Manchester, UK. Her main research interests include Simultaneous Inference, Multiple Comparison Procedures and Personalized Medicine. She received her PhD in Statistics from the University of Southampton in 2014.
Hope to see you at the webinar!