R. M. Crujeiras Casais, A. Meilan Vila, M. Francisco-Fernández
Climate change, although a global problem, is usually perceived by individuals on a local scale, being changes in temperature patterns the most direct indicator of global warning. Specifically, for an Atlantic climate location (as it happens in Galicia, NW-Spain), spring and fall are expected to present mild transitions between summer/winter, but people in this region usually perceive that this transition has disappeared in the last decade. We have considered daily temperature curves along time, being each curve attached to a certain calendar day, and therefore enabling the formulation of a regression model with functional covariate (the temperature curve) and circular response (the calendar day). Such a model is fitted with observations from the early 21st century. For observations from recent years, the fitted model is used to predict the calendar day, observing a remarkable misalignment between observed and predicted days for a certain curve.
Palabras clave: Circular regression, functional data, temperature curves
Programado
SGAPEIO-SEIO Sesión Invitada
8 de noviembre de 2023 10:10
CC2: Sala Conferencias