stilt is an open-source R library for Separable Gaussian Process Interpolation, a type of statistical emulation.
The package, which is available on CRAN, comprises a set of R functions to build and use an interpolator (emulator) for time series or 1D regularly spaced data in multidimensional space. The standard usage is for interpolating time-resolved computer model output between model parameter settings. It can also be used for interpolating multivariate data (e.g., oceanographic time-series data, etc.) in space. There are functions to test the emulator using cross-validation, and to produce contour plots over 2D slices in model input parameter (or physical) space.
To install the stable version from CRAN, simply run the following R code:
To install the latest development version directly from Github, run devtools:
install.packages("devtools") ##if not already installed library(devtools) devtools::install_github("scrim-network/Stilt-Rpkge") library(stilt)
To learn more we highly recommend you check out the reference manual. The reference manual explains each stilt function/dataset in-depth and includes annotated examples.
Additionally we recommend you check out the technical report. This report describes the statistical background and equations for the separable emulator.
R. Olson and W. Chang (2013): Mathematical framework for a separable Gaussian Process Emulator, Technical Report
stilt depends on R(≥2.15.1), fields
Roman Olson, Won Chang, Klaus Keller, and Murali Haran
Maintainer and Contact
Kelsey Ruckert: email@example.com
stilt is licensed under the GNU General Public License, version 3 or later.
Copyright 2013 Roman Olson, Won Chang, Klaus Keller, and Murali Haran
stilt is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
stilt is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. In addition, the authors and maintainer maintain no responsibility for any errors or bugs in the code.
This work was supported by the National Science Foundation through the Network for Sustainable Climate Risk Management (SCRiM) under NSF cooperative agreement GEO-1240507.