Package: mcgf 1.2.0.9000

mcgf: Markov Chain Gaussian Fields Simulation and Parameter Estimation

Simulating and estimating (regime-switching) Markov chain Gaussian fields with spatio-temporal covariance functions of the Gneiting class (Gneiting 2002) <doi:10.1198/016214502760047113>, including the regime-switching framework of Jia and Sezer (2025) <doi:10.1063/5.0285012>. It supports parameter estimation by weighted least squares and approximate conditional maximum likelihood methods, and produces Kriging forecasts and intervals for existing and new locations.

Authors:Tianxia Jia [aut, cre, cph]

mcgf_1.2.0.9000.tar.gz
mcgf_1.2.0.9000.zip(r-4.7-any)mcgf_1.2.0.9000.zip(r-4.6-any)mcgf_1.2.0.9000.zip(r-4.5-any)
mcgf_1.2.0.9000.tgz(r-4.6-any)mcgf_1.2.0.9000.tgz(r-4.5-any)
mcgf_1.2.0.9000.tar.gz(r-4.7-any)mcgf_1.2.0.9000.tar.gz(r-4.6-any)
mcgf_1.2.0.9000.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
mcgf/json (API)

# Install 'mcgf' in R:
install.packages('mcgf', repos = c('https://tianxia-jia.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/tianxia-jia/mcgf/issues

Pkgdown/docs site:https://tianxia-jia.github.io

Datasets:
  • sim1 - Simulated Markov chain Gaussian field
  • sim2 - Simulated regime-switching Markov chain Gaussian field
  • sim3 - Simulated regime-switching Markov chain Gaussian field
  • wind - Ireland wind data, 1961-1978

On CRAN:

Conda:

4.74 score 1 stars 22 scripts 311 downloads 44 exports 3 dependencies

Last updated from:c10aeafe24. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-develOK145
source / vignettesOK210
linux-releaseOK150
macos-releaseOK208
macos-oldrelOK232
windows-develOK110
windows-releaseOK86
windows-oldrelOK100
wasm-releaseOK100

Exports:acf_rsacfsacfs<-add_acfsadd_baseadd_ccfsadd_lagras.mcgf_rsbase<-ccf_rsccfsccfs<-ccovcor_cauchycor_expcor_fscor_lagr_askeycor_lagr_expcor_lagr_tricor_sepcor_statcor_stat_rscor2covcor2cov_ardistsdists<-find_distsfind_dists_newfit_basefit_lagris.mcgfis.mcgf_rskrigekrige_newlagr<-mcgfmcgf_rsmcgf_rs_simmcgf_simmodelrdistssd_rssdssds<-

Dependencies:latticeMASSsp

Choosing and Building Covariance Models
Why this vignette? | [C(\mathbf h,u) | Pure spatial and temporal kernels | Separable covariance | Fully symmetric Gneiting model | Lagrangian covariance functions | Triangular | Askey | Exponential | Combine the base and Lagrangian terms | A practical model-selection sequence | Parameter interpretation | References
Rendered fromcorrelation-models.Rmdusingknitr::rmarkdown

Last update: 2026-08-18
Started: 2026-08-18

Forecasting at New Locations
Overview | Prepare an MCGF object | Fit the base covariance model | Define a new location | Forecast with krige_new() | Forecast using new observed data | If observations at the new location are available | Supplying distances directly | General stationary model | Regime-switching objects
Rendered fromforecasting-new-locations.Rmdusingknitr::rmarkdown

Last update: 2026-08-18
Started: 2026-08-18

Steps for fitting an mcgf object
Introduction | Irish Wind | Data De-trending | Fitting Covariance Models | Pure Spatial Model | Pure Temporal Model | Separable Model | Fully Symmetric Model | Lagrangian Model | Simple Kriging | RMSE | MAE | POPI | Simple Kriging for new locations | References
Rendered frommcgf.Rmdusingknitr::rmarkdown

Last update: 2026-08-18
Started: 2023-10-30

Steps for fitting an mcgf_rs object
Introduction | Simulated Example | Regime process | Data Simulation | Locations | Data simulation | Fitting Covariance Models | Pure Spatial Model | Pure Temporal Model | Separable Model | Lagrangian Model | Simple Kriging for new locations | RMSE | MAE | POPI | References
Rendered frommcgf_rs.Rmdusingknitr::rmarkdown

Last update: 2026-08-18
Started: 2024-03-27

Simulating MCGF and RS-MCGF Processes
Overview | Distances | Simulate an MCGF | Add a Lagrangian component | Simulate an RS-MCGF | Returning covariance details | Common choices | References
Rendered fromsimulation.Rmdusingknitr::rmarkdown

Last update: 2026-08-18
Started: 2026-08-18

Readme and manuals

Help Manual

Help pageTopics
Calculate regime-switching auto-correlationacf_rs
Generic function for calculating autocorrelationacfs
Extract, calculate, or assign mean auto-correlations for an 'mcgf' or 'mcgf_rs' objectacfs.mcgf acfs.mcgf_rs acfs<- add_acfs
Generic function for adding a base modeladd_base
Add base model outputted from 'fit_base()' to an 'mcgf' object.add_base.mcgf base<-
Add base model outputted from 'fit_base()' to an 'mcgf_rs' object.add_base.mcgf_rs
Generic function for adding a Lagrangian modeladd_lagr
Add lagr model outputted from 'fit_lagr()' to a 'mcgf' object.add_lagr.mcgf lagr<-
Add lagr model outputted from 'fit_lagr()' to a 'mcgf_rs' object.add_lagr.mcgf_rs
Calculate regime-switching cross-correlationccf_rs
Generic function for calculating cross-correlationccfs
Extract, calculate, or assign cross-correlations for an 'mcgf' or 'mcgf_rs' objectadd_ccfs ccfs.mcgf ccfs.mcgf_rs ccfs<-
Generic functions for calculating joint covariance/correlation matrix for mcgf objectsccov
Covariance/correlation for joint distribution of an 'mcgf' objectccov.mcgf
Covariance/correlation for joint distribution of an 'mcgf_rs'objectccov.mcgf_rs
Calculate Cauchy correlationcor_cauchy
Calculate exponential correlationcor_exp
Calculate correlation for fully symmetric modelcor_fs
Calculate Lagrangian correlation of the Askey formcor_lagr_askey
Calculate Lagrangian correlation of the exponential formcor_lagr_exp
Calculate Lagrangian correlation of the triangular formcor_lagr_tri
Calculate correlation for separable modelcor_sep
Calculate general stationary correlation.cor_stat
Calculate general stationary correlation.cor_stat_rs
Convert correlation to covariancecor2cov cor2cov_ar
Generic function for calculating distance matricesdists
Calculating distance matrices for an 'mcgf' objectdists.mcgf dists<- `dists<-`
Calculate (signed) distances between coordinatesfind_dists
Calculate (signed) distances between coordinatesfind_dists_new
Fit correlation base modelsfit_base
Parameter estimation for symmetric correlation functions for an 'mcgf' object.fit_base.mcgf
Parameter estimation for symmetric correlation functions for an 'mcgf_rs' object.fit_base.mcgf_rs
Fit correlation Lagrangian modelsfit_lagr
Parameter estimation for Lagrangian correlation functions for an 'mcgf' object.fit_lagr.mcgf
Parameter estimation for Lagrangian correlation functions for an 'mcgf_rs' object.fit_lagr.mcgf_rs
Check if an object is an 'mcgf' object.is.mcgf
Check if an object is an 'mcgf_rs' object.as.mcgf_rs is.mcgf_rs
Generic function for computing kriging forecastskrige
Generic function for computing kriging forecasts for new locationskrige_new
Obtain kriging forecasts for new locations for an 'mcgf' object.krige_new.mcgf
Obtain kriging forecasts for new locations for an 'mcgf_rs' object.krige_new.mcgf_rs
Obtain kriging forecasts for an 'mcgf' object.krige.mcgf
Obtain kriging forecasts for an 'mcgf_rs' object.krige.mcgf_rs
Create mcgf objectmcgf
Create mcgf_rs objectmcgf_rs
Simulate regime-switching Markov chain Gaussian fieldmcgf_rs_sim
Simulate Markov chain Gaussian fieldmcgf_sim
Generic function for displaying fitted models for 'mcgf' objectsmodel
Display fitted models for an 'mcgf' or 'mcgf_rs' objectmodel.mcgf model.mcgf_rs
Print an 'mcgf' object.print.mcgf
Generate random distance matricesrdists
Calculate standard deviation for each location under each regime.sd_rs
Generic function for standard deviations for each columnsds
Extract, calculate, or assign standard deviations for an 'mcgf' or 'mcgf_rs' object.sds.mcgf sds.mcgf_rs sds<-
Simulated Markov chain Gaussian fieldsim1
Simulated regime-switching Markov chain Gaussian fieldsim2
Simulated regime-switching Markov chain Gaussian fieldsim3
Ireland wind data, 1961-1978wind