Stilt: Easy emulation of time series AR(1) computer model output in multidimensional parameter space

Roman Olson; Kelsey L. Ruckert; Won Chang; Klaus Keller; Murali Haran; Soon Il An

doi:10.32614/RJ-2018-049

Stilt: Easy emulation of time series AR(1) computer model output in multidimensional parameter space

Roman Olson, Kelsey L. Ruckert, Won Chang, Klaus Keller, Murali Haran, Soon Il An

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

Abstract

Statistically approximating or "emulating" time series model output in parameter space is a common problem in climate science and other fields. There are many packages for spatio-temporal modeling. However, they often lack focus on time series, and exhibit statistical complexity. Here, we present the R package stilt designed for simplified AR(1) time series Gaussian process emulation, and provide examples relevant to climate modelling. Notably absent is Markov chain Monte Carlo estimation - a challenging concept to many scientists. We keep the number of user choices to a minimum. Hence, the package can be useful pedagogically, while still applicable to real life emulation problems. We provide functions for emulator cross-validation, empirical coverage, prediction, as well as response surface plotting. While the examples focus on climate model emulation, the emulator is general and can be also used for kriging spatio-temporal data.

Original language	English
Pages (from-to)	209-225
Number of pages	17
Journal	R Journal
Volume	10
Issue number	2
DOIs	https://doi.org/10.32614/RJ-2018-049
Publication status	Published - 2019

Bibliographical note

Funding Information:
For their roles in producing, coordinating, and making available the CMIP5 model output, we acknowledge the climate modeling groups, the World Climate Research Programme's (WCRP) Working Group on Coupled modeling (WGCM), and the Global Organization for Earth System Science Portals (GO-ESSP). We thank Jong-Soo Shin for help with extracting Korean temperature output, and Patrick Applegate for sharing the SICOPOLIS ice sheet model output. We acknowledge financial support from National Research Foundation of Korea (NRF-2009-0093069, NRF-2018R1A5A1024958), and from the Institute for Basic Science (project code IBS-R028-D1). This work was also co-supported by the National Science Foundation through the Network for Sustainable Climate Risk Management (SCRiM) under NSF cooperative agreement GEO-1240507 and the Penn State Center for Climate Risk Management. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, or any other foundation or entity

Publisher Copyright:
© 2018 The R Journal.

All Science Journal Classification (ASJC) codes

Statistics and Probability
Numerical Analysis
Statistics, Probability and Uncertainty

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.32614/RJ-2018-049

Cite this

@article{e276952555a94822b196152594601339,

title = "Stilt: Easy emulation of time series AR(1) computer model output in multidimensional parameter space",

abstract = "Statistically approximating or {"}emulating{"} time series model output in parameter space is a common problem in climate science and other fields. There are many packages for spatio-temporal modeling. However, they often lack focus on time series, and exhibit statistical complexity. Here, we present the R package stilt designed for simplified AR(1) time series Gaussian process emulation, and provide examples relevant to climate modelling. Notably absent is Markov chain Monte Carlo estimation - a challenging concept to many scientists. We keep the number of user choices to a minimum. Hence, the package can be useful pedagogically, while still applicable to real life emulation problems. We provide functions for emulator cross-validation, empirical coverage, prediction, as well as response surface plotting. While the examples focus on climate model emulation, the emulator is general and can be also used for kriging spatio-temporal data.",

author = "Roman Olson and Ruckert, {Kelsey L.} and Won Chang and Klaus Keller and Murali Haran and An, {Soon Il}",

note = "Funding Information: For their roles in producing, coordinating, and making available the CMIP5 model output, we acknowledge the climate modeling groups, the World Climate Research Programme's (WCRP) Working Group on Coupled modeling (WGCM), and the Global Organization for Earth System Science Portals (GO-ESSP). We thank Jong-Soo Shin for help with extracting Korean temperature output, and Patrick Applegate for sharing the SICOPOLIS ice sheet model output. We acknowledge financial support from National Research Foundation of Korea (NRF-2009-0093069, NRF-2018R1A5A1024958), and from the Institute for Basic Science (project code IBS-R028-D1). This work was also co-supported by the National Science Foundation through the Network for Sustainable Climate Risk Management (SCRiM) under NSF cooperative agreement GEO-1240507 and the Penn State Center for Climate Risk Management. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, or any other foundation or entity Publisher Copyright: {\textcopyright} 2018 The R Journal.",

year = "2019",

doi = "10.32614/RJ-2018-049",

language = "English",

volume = "10",

pages = "209--225",

journal = "R Journal",

issn = "2073-4859",

publisher = "R Foundation for Statistical Computing",

number = "2",

}

TY - JOUR

T1 - Stilt

T2 - Easy emulation of time series AR(1) computer model output in multidimensional parameter space

AU - Olson, Roman

AU - Ruckert, Kelsey L.

AU - Chang, Won

AU - Keller, Klaus

AU - Haran, Murali

AU - An, Soon Il

N1 - Funding Information: For their roles in producing, coordinating, and making available the CMIP5 model output, we acknowledge the climate modeling groups, the World Climate Research Programme's (WCRP) Working Group on Coupled modeling (WGCM), and the Global Organization for Earth System Science Portals (GO-ESSP). We thank Jong-Soo Shin for help with extracting Korean temperature output, and Patrick Applegate for sharing the SICOPOLIS ice sheet model output. We acknowledge financial support from National Research Foundation of Korea (NRF-2009-0093069, NRF-2018R1A5A1024958), and from the Institute for Basic Science (project code IBS-R028-D1). This work was also co-supported by the National Science Foundation through the Network for Sustainable Climate Risk Management (SCRiM) under NSF cooperative agreement GEO-1240507 and the Penn State Center for Climate Risk Management. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, or any other foundation or entity Publisher Copyright: © 2018 The R Journal.

PY - 2019

Y1 - 2019

N2 - Statistically approximating or "emulating" time series model output in parameter space is a common problem in climate science and other fields. There are many packages for spatio-temporal modeling. However, they often lack focus on time series, and exhibit statistical complexity. Here, we present the R package stilt designed for simplified AR(1) time series Gaussian process emulation, and provide examples relevant to climate modelling. Notably absent is Markov chain Monte Carlo estimation - a challenging concept to many scientists. We keep the number of user choices to a minimum. Hence, the package can be useful pedagogically, while still applicable to real life emulation problems. We provide functions for emulator cross-validation, empirical coverage, prediction, as well as response surface plotting. While the examples focus on climate model emulation, the emulator is general and can be also used for kriging spatio-temporal data.

AB - Statistically approximating or "emulating" time series model output in parameter space is a common problem in climate science and other fields. There are many packages for spatio-temporal modeling. However, they often lack focus on time series, and exhibit statistical complexity. Here, we present the R package stilt designed for simplified AR(1) time series Gaussian process emulation, and provide examples relevant to climate modelling. Notably absent is Markov chain Monte Carlo estimation - a challenging concept to many scientists. We keep the number of user choices to a minimum. Hence, the package can be useful pedagogically, while still applicable to real life emulation problems. We provide functions for emulator cross-validation, empirical coverage, prediction, as well as response surface plotting. While the examples focus on climate model emulation, the emulator is general and can be also used for kriging spatio-temporal data.

UR - http://www.scopus.com/inward/record.url?scp=85070269481&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070269481&partnerID=8YFLogxK

U2 - 10.32614/RJ-2018-049

DO - 10.32614/RJ-2018-049

M3 - Article

AN - SCOPUS:85070269481

SN - 2073-4859

VL - 10

SP - 209

EP - 225

JO - R Journal

JF - R Journal

IS - 2

ER -

Stilt: Easy emulation of time series AR(1) computer model output in multidimensional parameter space

Abstract

Bibliographical note

All Science Journal Classification (ASJC) codes

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this