Statistical Modelling 5 (2005), 289–307
Understanding past ocean circulations: a
nonparametric regression case study
Richard Samworth
Statslab, Centre for Mathematical Sciences,
Wilberforce Road,
Cambridge CB3 0WB
UK
eMail: rjs57@can.ac.uk
Heather Poore
Department of Earth Sciences, Bullard Laboratories,
Madingley Road,
Cambridge CB3 0EZ
Abstract:
Oceanographers study past ocean circulations and their effect on
global climate through carbon
isotope records obtained from microfossils deposited on the
ocean floor. An initial goal is to estimate the
carbon isotope levels for the Pacific, Southern and North
Atlantic Oceans over the last 23 million years and
to provide confidence bands.We consider a nonparametric
regression model and demonstrate how several
recent developments in methodology make local linear kernel
regression an attractive approach for tackling
the problem. The results are used to estimate a quantity
called the proportion of Northern Component
Water and its effect on global climate. Several interesting
and important geophysical and oceanographic
conclusions are suggested by the study.
Keywords:
confidence bands; errors-in-variables;
local linear kernel estimator; model checking;
nonparametric regression; Northern Component Water;
SIMEX algorithm
Downloads:
Data
in zipped archive
Color plot in pdf format
back