<mods:mods xmlns:mods="http://www.loc.gov/mods/v3" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-7.xsd"><mods:titleInfo><mods:title>State-space Models and Gaussian Processes in Paleoceanography and Paleoclimatology</mods:title></mods:titleInfo><mods:name type="personal"><mods:namePart>Lee, Taehee</mods:namePart><mods:role><mods:roleTerm type="text">creator</mods:roleTerm></mods:role></mods:name><mods:name type="personal"><mods:namePart>Lawrence, Charles</mods:namePart><mods:role><mods:roleTerm type="text">Advisor</mods:roleTerm></mods:role></mods:name><mods:name type="personal"><mods:namePart>Lisiecki, Lorraine</mods:namePart><mods:role><mods:roleTerm type="text">Reader</mods:roleTerm></mods:role></mods:name><mods:name type="personal"><mods:namePart>Harrison, Matthew</mods:namePart><mods:role><mods:roleTerm type="text">Reader</mods:roleTerm></mods:role></mods:name><mods:name type="corporate"><mods:namePart>Brown University. Department of Applied Mathematics</mods:namePart><mods:role><mods:roleTerm type="text">sponsor</mods:roleTerm></mods:role></mods:name><mods:originInfo><mods:copyrightDate>2020</mods:copyrightDate></mods:originInfo><mods:physicalDescription><mods:extent>xiv, 186 p.</mods:extent><mods:digitalOrigin>born digital</mods:digitalOrigin></mods:physicalDescription><mods:note type="thesis">Thesis (Ph. D.)--Brown University, 2020</mods:note><mods:genre authority="aat">theses</mods:genre><mods:abstract>For the recent decades, statistical (machine) learning and Bayesian modelling have emerged as popular frameworks for various fields of area, regardless of academy or industry. The statistical learning theory has many applicable branches that include state-space models and the Gaussian process. State-space models are designed to deal with latent series under the Markov assumption. The hidden Markov model is applied for the case that each hidden state is defined on a finite set, whereas the Kalman filter, particle filter and Markov-chain Monte Carlo algorithms can treat the continuous cases. The Gaussian process gives a nonparametric prior on the functions. Its flexibility allows various applications including the regression, classification, state-space models and dimensionality reduction.&#13;
Paleoceanography and paleoclimatology are the fields of geoscience that discover the past ocean and climate events from the relevant preserved physical characteristics, proxies, as well as their ages. Interdisciplinary research encompassing physics, chemistry and biology has investigated the relationship between ocean and climate events and proxies, and more recently statistical learning is also making progresses. The scarce and agnostic characteristics of data need particular modelling that are more dependent on the data than certain parametric structures.&#13;
In this thesis, three applications of the state-space models and Gaussian process to the field of paleoceanography and paleoclimatology are presented. Dual proxy stack construction algorithm addresses the age assignments problem with the radiocarbon (14C) and benthic δ18O proxies by adopting a state-space model and the Gaussian process regression. The heteroscedastic Gaussian process regression constructs nonparametric calibration models of alkenone and TEX86 proxies for restoring past sea surface temperatures and a novel multimodal Gaussian process regression model is also considered. Two state-space models are applied for reconstructing past atmospheric CO2 pressures with the boron isotope and benthic δ18O proxies: one is a parametric state-space model based on the particle smoother and the other is a nonparametric Gaussian process state-space model. Our study certifies the effectiveness of giving correlation to the hidden states and considering nonparametric data-driven models in paleoceanography and paleoclimatology.</mods:abstract><mods:subject authority="fast" authorityURI="http://id.worldcat.org/fast" valueURI="http://id.worldcat.org/fast/01400410"><mods:topic>Applied mathematics</mods:topic></mods:subject><mods:subject authority="fast" authorityURI="http://id.worldcat.org/fast" valueURI="http://id.worldcat.org/fast/01051364"><mods:topic>Paleoclimatology</mods:topic></mods:subject><mods:subject authority="fast" authorityURI="http://id.worldcat.org/fast" valueURI="http://id.worldcat.org/fast/00939020"><mods:topic>Gaussian processes</mods:topic></mods:subject><mods:subject><mods:topic>State-space models</mods:topic></mods:subject><mods:subject authority="fast" authorityURI="http://id.worldcat.org/fast" valueURI="http://id.worldcat.org/fast/01004801"><mods:topic>Machine learning--Statistical methods</mods:topic></mods:subject><mods:subject><mods:topic>Bayesian Machine Learning</mods:topic></mods:subject><mods:subject authority="fast" authorityURI="http://id.worldcat.org/fast" valueURI="http://id.worldcat.org/fast/01051357"><mods:topic>Paleoceanography</mods:topic></mods:subject><mods:language><mods:languageTerm authority="iso639-2b">English</mods:languageTerm></mods:language><mods:recordInfo><mods:recordContentSource authority="marcorg">RPB</mods:recordContentSource><mods:recordCreationDate encoding="iso8601">20200720</mods:recordCreationDate></mods:recordInfo><mods:identifier type="doi">10.26300/etmx-ty09</mods:identifier><mods:accessCondition type="rights statement" xlink:href="http://rightsstatements.org/vocab/InC/1.0/">In Copyright</mods:accessCondition><mods:accessCondition type="restriction on access">All rights reserved. Collection is open to the Brown community for research.</mods:accessCondition><mods:typeOfResource authority="primo">dissertations</mods:typeOfResource></mods:mods>