psfmi: Prediction Model Selection and Performance Evaluation in Multiple Imputed Datasets

Pooling, backward and forward selection of logistic and Cox regression models in multiply imputed datasets. Backward and forward selection can be done from the pooled model using Rubin's Rules (RR), the D1, D2, D3 and the median p-values method. This is also possible for Mixed models. The models can contain continuous, dichotomous, categorical and restricted cubic spline predictors and interaction terms between all these type of predictors. The stability of the models can be evaluated using bootstrapping and cluster bootstrapping. The package further contains functions to pool the model performance as ROC/AUC, R-squares, scaled Brier score and calibration plots for logistic regression models. Internal validation can be done with cross-validation or bootstrapping. The adjusted intercept after shrinkage of pooled regression coefficients can be obtained. Backward and forward selection as part of internal validation is possible. A function to externally validate logistic prediction models in multiple imputed datasets is available and a function to compare models. Eekhout (2017) <doi:10.1186/s12874-017-0404-7>. Wiel (2009) <doi:10.1093/biostatistics/kxp011>. Marshall (2009) <doi:10.1186/1471-2288-9-57>.

Version: 0.7.1
Depends: R (≥ 4.0.0)
Imports: ggplot2 (≥ 3.3.2), norm (≥ 1.0-9.5), survival (≥ 3.1-12), mitools (≥ 2.4), pROC (≥ 1.16.2), rms (≥ 6.1-0), ResourceSelection (≥ 0.3-5), magrittr (≥ 2.0.1), rsample (≥ 0.0.8), mice (≥ 3.12.0), mitml (≥ 0.3-7), cvAUC (≥ 1.1.0), dplyr (≥ 1.0.2), purrr (≥ 0.3.4), tidyr (≥ 1.1.2), tibble (≥ 3.0.4), stringr (≥ 1.4.0), lme4 (≥ 1.1-26), miceadds (≥ 3.10-28), car (≥ 3.0-10)
Suggests: foreign (≥ 0.8-80), knitr, rmarkdown, testthat, bookdown, readr
Published: 2021-01-13
Author: Martijn Heymans ORCID iD [cre, aut], Iris Eekhout [ctb]
Maintainer: Martijn Heymans <mw.heymans at amsterdamumc.nl>
BugReports: https://github.com/mwheymans/psfmi/issues/
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL: https://mwheymans.github.io/psfmi/
NeedsCompilation: no
Materials: README NEWS
In views: MissingData
CRAN checks: psfmi results

Downloads:

Reference manual: psfmi.pdf
Vignettes: Multiple Imputation and Bootstrapping - Method MI_boot
Multiple Imputation and Cross-validation - The MI_cv_naive method
Multiple Imputation and Bootstrapping - Method boot_MI
Multiple Imputation and Cross-validation - method cv_MI
Multiple Imputation and Cross-validation - Method cv_MI_RR
Workflow to Develop and Validate a Prediction model in Multiply Imputed data
Pooling and Selection of Cox Regression Models
Pooling and Selection of Logistic Regression Models
Stability analysis after Multiple Imputation
Working together: mice and psfmi
Package source: psfmi_0.7.1.tar.gz
Windows binaries: r-devel: psfmi_0.7.1.zip, r-release: psfmi_0.5.0.zip, r-oldrel: psfmi_0.5.0.zip
macOS binaries: r-release: psfmi_0.7.1.tgz, r-oldrel: psfmi_0.5.0.tgz
Old sources: psfmi archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=psfmi to link to this page.