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 journalArticle

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 languageEnglish (US)
Pages (from-to)209-225
Number of pages17
JournalR Journal
Volume10
Issue number2
DOIs
StatePublished - Jan 1 2019

Fingerprint

Emulation
Computer Model
Parameter Space
Time series
Output
Climate Modeling
Spatio-temporal Modeling
Climate models
Spatio-temporal Data
Climate Models
Response Surface
Kriging
Time Series Models
Markov Chain Monte Carlo
Gaussian Process
Cross-validation
Climate
Markov processes
Coverage
Prediction

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

Cite this

Olson, Roman ; Ruckert, Kelsey L. ; Chang, Won ; Keller, Klaus ; Haran, Murali ; An, Soon Il. / Stilt : Easy emulation of time series AR(1) computer model output in multidimensional parameter space. In: R Journal. 2019 ; Vol. 10, No. 2. pp. 209-225.
@article{37998f6b02544578856595c99c271568,
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}",
year = "2019",
month = "1",
day = "1",
doi = "10.32614/RJ-2018-049",
language = "English (US)",
volume = "10",
pages = "209--225",
journal = "R Journal",
issn = "2073-4859",
publisher = "R Foundation for Statistical Computing",
number = "2",

}

Stilt : Easy emulation of time series AR(1) computer model output in multidimensional parameter space. / Olson, Roman; Ruckert, Kelsey L.; Chang, Won; Keller, Klaus; Haran, Murali; An, Soon Il.

In: R Journal, Vol. 10, No. 2, 01.01.2019, p. 209-225.

Research output: Contribution to journalArticle

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

PY - 2019/1/1

Y1 - 2019/1/1

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

VL - 10

SP - 209

EP - 225

JO - R Journal

JF - R Journal

SN - 2073-4859

IS - 2

ER -