A functional-circular analysis of temperature curves
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.
Keywords: Circular regression, functional data, temperature curves
Scheduled
SGAPEIO-SEIO Invited Session
November 8, 2023 10:10 AM
CC2: Conference Room
Other papers in the same session
M. J. Ginzo Villamayor, P. Saavedra Nieves, D. Royé, F. Caamaño Isorna
N. Diz-Rosales, M. J. Lombardía, D. Morales
W. González Manteiga, M. I. Borrajo García, I. Fuentes Santos