Department of Mathematical Sciences, The University of Liverpool, Liverpool L69 7ZL, U.K.
Abstract:
We consider a local least-squares criterion for aligning multipletime series fragments differing by locations and show the consistencyof the time-lag estimator and the asymptotic normality of thelocation estimator. We apply the criterion to the problem ofaligning 50 glacial varve fragments and construct a 3000-yearsurrogate for global temperature.