Cross-validation | Examples | Polynomial regression for the Auto data | Using cv() | Comparing competing models | Logistic regression for the Mroz data | Replicating cross-validation | Manipulating "cv" and related objects | Parallel computations | References
Package: cv 2.0.6
Georges Monette
cv: Cross-Validating Regression Models
Cross-validation methods of regression models that exploit features of various modeling functions to improve speed. Some of the methods implemented in the package are novel, as described in Fox and Monette (2026) <doi:10.18637/jss.v116.i08>, and the package vignettes. For general introductions to cross-validation, see, for example, Gareth James, Daniela Witten, Trevor Hastie, and Robert Tibshirani (2021, ISBN 978-1-0716-1417-4, Secs. 5.1, 5.3), "An Introduction to Statistical Learning with Applications in R, Second Edition", and Trevor Hastie, Robert Tibshirani, and Jerome Friedman (2009, ISBN 978-0-387-84857-0, Sec. 7.10), "The Elements of Statistical Learning, Second Edition".
Authors:
cv_2.0.6.tar.gz
cv_2.0.6.zip(r-4.7)cv_2.0.6.zip(r-4.6)cv_2.0.6.zip(r-4.5)
cv_2.0.6.tgz(r-4.6-any)cv_2.0.6.tgz(r-4.5-any)
cv_2.0.6.tar.gz(r-4.7-any)cv_2.0.6.tar.gz(r-4.6-any)
cv_2.0.6.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
cv/json (API)
| # Install 'cv' in R: |
| install.packages('cv', repos = c('https://gmonette.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/gmonette/cv/issues
Pkgdown/docs site:https://gmonette.github.io
Datasets:
- Pigs - Body Weights of 48 Pigs in 9 Successive Weeks
Last updated from:7facc7a9f2. Checks:9 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel-x86_64 | OK | 191 | ||
| source / vignettes | OK | 469 | ||
| linux-release-x86_64 | OK | 195 | ||
| macos-release-arm64 | OK | 109 | ||
| macos-oldrel-arm64 | OK | 114 | ||
| windows-devel | OK | 131 | ||
| windows-release | OK | 114 | ||
| windows-oldrel | OK | 129 | ||
| wasm-release | OK | 132 |
Exports:BayesRuleBayesRule2checkFormulacompareFoldscvcvComputecvInfocvMixedcvSelectfoldfoldsGetResponsemedAbsErrmodelsmsermseselectModelListselectStepAICselectTransselectTransStepAIC
Dependencies:abindbackportsbootbroomcarcarDataclicodetoolscolorspacecowplotcpp11DerivdoBydoParalleldplyrfarverforeachforecastFormulafracdiffgenericsggplot2glmmTMBgluegtablegtoolsinsightisobanditeratorslabelinglatticelifecyclelme4lmtestmagrittrMASSMatrixMatrixModelsmgcvmicrobenchmarkminqamodelrnlmenloptrnnetnumDerivpbkrtestpillarpkgconfigpurrrquantregR6rbibutilsRColorBrewerRcppRcppArmadilloRcppEigenRdpackreformulasrlangS7sandwichscalesSparseMstringistringrsurvivaltibbletidyrtidyselecttimeDateTMBurcautf8vctrsviridisLitewithrzoo
Cross-validating regression models
Last update: 2026-06-09
Started: 2023-08-03
Cross-validating model selection
A preliminary example | Polynomial regression for the Auto data revisited: meta cross-validation | Mroz's logistic regression revisited | Cross-validating choice of transformations in regression | Selecting both transformations and predictors[^Venables] | References
Last update: 2024-09-15
Started: 2024-04-03
Extending the cv package
Adding a cost criterion | Adding a model class not covered by the default cv() method | Independently sampled cases | Calling cvCompute() | Mixed-effects models | Adding a model-selection procedure | References
Last update: 2024-09-05
Started: 2023-08-25
Computational and technical notes on cross-validating regression models
Efficient computations for linear and generalized linear models | Computation of the bias-corrected CV criterion and confidence intervals | Why the complement of AUC isn't a casewise CV criterion | The values of $\mathrm{cv}(\widehat{y}_i, y_i)$ that express AUC as a sum of casewise values are solutions of equation (\ref{eq:cw}), which can be written as solutions of the following system of $2^{n_v}$ linear simultaneous equations in $2n_v$ unknowns:\begin{equation}\label{eq:lin}\tag{2}(\mathbf{U} -\mathbf{Y}) \mathbf{c}_0 + \mathbf{Y} \mathbf{c}_1 | References
Last update: 2024-09-02
Started: 2024-04-03
Cross-validating mixed-effects models
Example: The High-School and Beyond data | Example: Contrasting cluster-based and case-based CV | Example: Crossed random effects | References
Last update: 2024-09-02
Started: 2024-04-03
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Cross-Validate Regression Models | as.data.frame.cv as.data.frame.cvList cv cv.default cv.glm cv.lm cv.rlm cvInfo cvInfo.cv cvInfo.cvList cvInfo.cvModList plot.cv plot.cvList print.cv print.cvDataFrame print.cvList summary.cv summary.cvDataFrame summary.cvList |
| Cross-Validate a Model-Selection Procedure | coef.cvSelect compareFolds cv.function cvInfo.cvSelect selectModelList selectStepAIC selectTrans selectTransStepAIC |
| Cross-Validate Mixed-Effects Model | cv.glmmTMB cv.lme cv.merMod |
| Cross-Validate Several Models Fit to the Same Data | as.data.frame.cvModList cv.modList models plot.cvModList print.cvModList summary.cvModList |
| Utility Functions for the cv Package | checkFormula cvCompute cvMixed cvSelect fold fold.folds folds GetResponse GetResponse.default GetResponse.glmmTMB GetResponse.lme GetResponse.merMod GetResponse.modList print.folds |
| Cost Functions for Fitted Regression Models | BayesRule BayesRule2 costFunctions medAbsErr mse rmse |
| Body Weights of 48 Pigs in 9 Successive Weeks | Pigs |