# Partial Least Squares for Serially Dependent Data

 dc.contributor.advisor Krivobokova, Tatyana Prof. Dr. dc.contributor.author Singer, Marco dc.date.accessioned 2016-09-12T09:47:33Z dc.date.available 2016-09-12T09:47:33Z dc.date.issued 2016-09-12 dc.identifier.uri http://hdl.handle.net/11858/00-1735-0000-0028-8831-B dc.identifier.uri http://dx.doi.org/10.53846/goediss-5828 dc.language.iso deu de dc.rights.uri http://creativecommons.org/licenses/by-nc-nd/4.0/ dc.subject.ddc 510 de dc.title Partial Least Squares for Serially Dependent Data de dc.type doctoralThesis de dc.contributor.referee Krivobokova, Tatyana Prof. Dr. dc.date.examination 2016-08-04 dc.description.abstracteng In the first paper we consider the partial least squares algorithm for dependent data and study the consequences de of ignoring the dependence both theoretically and numerically. Ignoring nonstationary dependence structures can lead to inconsistent estimation, but a simple modification leads to consistent estimation. A protein dynamics example illustrates the superior predictive power of the method. For the second paper we consider the kernel partial least squares algorithm for the solution of nonparametric regression problems when the data exhibit dependence in their observations in the form of stationary time series. Probabilistic convergence rates of the kernel partial least squares estimator to the true regression function are established under a source condition. The impact of long range dependence in the data is studied both theoretically and in simulations. dc.contributor.coReferee Munk, Axel Prof. Dr. dc.subject.eng Dependent data, Kernel partial least squares, Latent variable model, Long range dependence, Nonparametric regression, Nonstationary process, Partial least squares, Protein dynamics, Source condition, Stationary process de dc.identifier.urn urn:nbn:de:gbv:7-11858/00-1735-0000-0028-8831-B-2 dc.affiliation.institute Fakultät für Mathematik und Informatik de dc.subject.gokfull Mathematik (PPN61756535X) de dc.identifier.ppn 869469967
﻿