Package: PDtoolkit 1.2.1

PDtoolkit: Collection of Tools for PD Rating Model Development and Validation

The goal of this package is to cover the most common steps in probability of default (PD) rating model development and validation. The main procedures available are those that refer to univariate, bivariate, multivariate analysis, calibration and validation. Along with accompanied 'monobin' and 'monobinShiny' packages, 'PDtoolkit' provides functions which are suitable for different data transformation and modeling tasks such as: imputations, monotonic binning of numeric risk factors, binning of categorical risk factors, weights of evidence (WoE) and information value (IV) calculations, WoE coding (replacement of risk factors modalities with WoE values), risk factor clustering, area under curve (AUC) calculation and others. Additionally, package provides set of validation functions for testing homogeneity, heterogeneity, discriminatory and predictive power of the model.

Authors:Andrija Djurovic [aut, cre]

PDtoolkit_1.2.1.tar.gz
PDtoolkit_1.2.1.zip(r-4.5)PDtoolkit_1.2.1.zip(r-4.4)PDtoolkit_1.2.1.zip(r-4.3)
PDtoolkit_1.2.1.tgz(r-4.4-any)PDtoolkit_1.2.1.tgz(r-4.3-any)
PDtoolkit_1.2.1.tar.gz(r-4.5-noble)PDtoolkit_1.2.1.tar.gz(r-4.4-noble)
PDtoolkit_1.2.1.tgz(r-4.4-emscripten)PDtoolkit_1.2.1.tgz(r-4.3-emscripten)
PDtoolkit.pdf |PDtoolkit.html✨
PDtoolkit/json (API)
NEWS

# Install 'PDtoolkit' in R:

install.packages('PDtoolkit', repos = c('https://andrija-djurovic.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/andrija-djurovic/pdtoolkit/issues

Datasets:

loans - German Credit Data

On CRAN:

4.78 score 14 stars 86 scripts 335 downloads 47 exports 69 dependencies

Last updated 10 months agofrom:d2013e3d3c. Checks:OK: 3 NOTE: 4. Indexed: yes.

Target	Result	Date
Doc / Vignettes	OK	Nov 21 2024
R-4.5-win	NOTE	Nov 21 2024
R-4.5-linux	NOTE	Nov 21 2024
R-4.4-win	NOTE	Nov 21 2024
R-4.4-mac	NOTE	Nov 21 2024
R-4.3-win	OK	Nov 21 2024
R-4.3-mac	OK	Nov 21 2024

Exports:auc.model bivariate boots.vld cat.bin cat.slice confusion.matrix constrained.logit create.partitions cutoff.palette decision.tree dp.testing embedded.blocks encode.woe ensemble.blocks evrs fairness.vld heterogeneity hhi homogeneity imp.outliers imp.sc interaction.transformer kfold.idx kfold.vld normal.test num.slice nzv power pp.testing psi replace.woe rf.clustering rf.interaction.transformer rs.calibration scaled.score segment.vld smote staged.blocks stepFWD stepFWDr stepMIV stepRPC stepRPCr univariate ush.bin ush.test woe.tbl

Dependencies:backports base64enc bslib cachem checkmate cli cluster colorspace data.table digest dplyr evaluate fansi farver fastmap fontawesome foreign Formula fs generics ggplot2 glue gridExtra gtable highr Hmisc htmlTable htmltools htmlwidgets isoband jquerylib jsonlite knitr labeling lattice lifecycle magrittr MASS Matrix memoise mgcv mime monobin munsell nlme nnet pillar pkgconfig R6 rappdirs RColorBrewer rlang rmarkdown rpart rstudioapi sass scales stringi stringr tibble tidyselect tinytex utf8 vctrs viridis viridisLite withr xfun yaml

Help page	Topics
Area under curve (AUC)	auc.model
Bivariate analysis	bivariate
Bootstrap model validation	boots.vld
Categorical risk factor binning	cat.bin
Slice categorical variable	cat.slice
Confusion matrix	confusion.matrix
Constrained logistic regression	constrained.logit
Create partitions (aka nested dummy variables)	create.partitions
Palette of cutoff values that minimize and maximize metrics from the confusion matrix	cutoff.palette
Custom decision tree algorithm	decision.tree
Testing the discriminatory power of PD rating model	dp.testing
Embedded blocks regression	embedded.blocks
Encode WoE	encode.woe
Ensemble blocks regression	ensemble.blocks
Modelling the Economic Value of Credit Rating System	evrs
Model fairness validation	fairness.vld
Testing heterogeneity of the PD rating model	heterogeneity
Herfindahl-Hirschman Index (HHI)	hhi
Testing homogeneity of the PD rating model	homogeneity
Imputation methods for outliers	imp.outliers
Imputation methods for special cases	imp.sc
Extract risk factors interaction from decision tree	interaction.transformer
Indices for K-fold validation	kfold.idx
K-fold model cross-validation	kfold.vld
German Credit Data	loans
Multi-period predictive power test	normal.test
Slice numeric variable	num.slice
Near-zero variance	nzv
Power of statistical tests for predictive ability testing	power
Testing the predictive power of PD rating model	pp.testing
Predict method for custom decision tree	predict.cdt
Population Stability Index (PSI)	psi
Replace modalities of risk factor with weights of evidence (WoE) value	replace.woe
Risk factor clustering	rf.clustering
Extract interactions from random forest	rf.interaction.transformer
Calibration of the rating scale	rs.calibration
Scaling the probabilities	scaled.score
Model segment validation	segment.vld
Synthetic Minority Oversampling Technique (SMOTE)	smote
Staged blocks regression	staged.blocks
Customized stepwise regression with p-value and trend check	stepFWD
Customized stepwise regression with p-value and trend check on raw risk factors	stepFWDr
Stepwise logistic regression based on marginal information value (MIV)	stepMIV
Stepwise logistic regression based on risk profile concept	stepRPC
Stepwise regression based on risk profile concept and raw risk factors	stepRPCr
Univariate analysis	univariate
U-shape binning algorithm	ush.bin
Testing for U-shape relation	ush.test
Weights of evidence (WoE) table	woe.tbl

Package: PDtoolkit 1.2.1

PDtoolkit: Collection of Tools for PD Rating Model Development and Validation

Citation

Development and contributors

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)