Create Credit Scorecards

Credit scorecard modeling, binning, fitting a model, obtaining points and scores, model validation, probability of default, create compact scorecard

For information about the workflow for developing credit scorecards, see Credit Scorecard Modeling Workflow. Additional tools for working with credit scorecards are available in Risk Management Toolbox™. For more information, see Consumer Credit Risk (Risk Management Toolbox).

Objects

creditscorecard Create creditscorecard object to build credit scorecard model

Functions

`autobinning`	Perform automatic binning of given predictors
`bininfo`	Return predictor’s bin information
`predictorinfo`	Summary of credit scorecard predictor properties
`fillmissing`	Replace missing values for credit scorecard predictors (Since R2020a)
`modifybins`	Modify predictor’s bins
`modifypredictor`	Set properties of credit scorecard predictors
`bindata`	Binned predictor variables
`plotbins`	Plot histogram counts for predictor variables
`fitmodel`	Fit logistic regression model to Weight of Evidence (WOE) data
`fitConstrainedModel`	Fit logistic regression model to Weight of Evidence (WOE) data subject to constraints on model coefficients
`setmodel`	Set model predictors and coefficients
`displaypoints`	Return points per predictor per bin
`formatpoints`	Format scorecard points and scaling
`score`	Compute credit scores for given data
`probdefault`	Likelihood of default for given data set
`validatemodel`	Validate quality of credit scorecard model
`screenpredictors`	Screen credit scorecard predictors for predictive value
`compact`	Create compact credit scorecard

Live Editor Tasks

Thresholds for Screen Predictors

Select thresholds for predictor risk metrics in the Live Editor (Since R2021b)

Topics

Feature Screening with screenpredictors (Risk Management Toolbox)
This example shows how to perform predictor screening using screenpredictors and then set predictor thresholds using the Threshold Predictors live task.
Case Study for Credit Scorecard Analysis
This example shows how to create a creditscorecard object, bin data, display, and plot binned data information.
Bin Data to Create Credit Scorecards Using Binning Explorer (Risk Management Toolbox)
Create a credit scorecard using the Binning Explorer app.
Credit Scorecards with Constrained Logistic Regression Coefficients
To compute scores for a creditscorecard object with constraints for equality, inequality, or bounds on the coefficients of the logistic regression model, use fitConstrainedModel.
Credit Scorecard Modeling with Missing Values
This example shows alternative workflows to handle missing values when working with creditscorecard objects.
Credit Scoring Using Logistic Regression and Decision Trees (Risk Management Toolbox)
Create and compare two credit scoring models, one based on logistic regression and the other based on decision trees.
Use Reject Inference Techniques with Credit Scorecards (Risk Management Toolbox)
This example demonstrates the hard-cutoff and fuzzy augmentation approaches to reject inference.
Credit Rating by Bagging Decision Trees
This example shows how to build an automated credit rating tool.
Credit Rating by Ordinal Multinomial Regression (Risk Management Toolbox)
This example shows how to use ordinal multinomial logistic regression to build a credit rating model that you can use in an automated credit rating process.
compactCreditScorecard Object Workflow (Risk Management Toolbox)
This example shows a workflow for creating a compactCreditScorecard object from a creditscorecard object.
Impute Missing Data in the Credit Scorecard Workflow Using the k-Nearest Neighbors Algorithm
This example shows how to perform imputation of missing data in the credit scorecard workflow using the k-nearest neighbors (kNN) algorithm.
Impute Missing Data in the Credit Scorecard Workflow Using the Random Forest Algorithm
This example shows how to perform imputation of missing data in the credit scorecard workflow using the random forest algorithm.
Treat Missing Data in a Credit Scorecard Workflow Using MATLAB fillmissing
This example shows a workflow to gather missing data, manually treat the training data, develop a new creditscorecard, and treat new data before scoring using the MATLAB® fillmissing.
Explore Fairness Metrics for Credit Scoring Model (Risk Management Toolbox)
Calculate and use data and model metrics to investigate the biases that exist in a model.
Bias Mitigation in Credit Scoring by Reweighting (Risk Management Toolbox)
Use bias mitigation with a credit scorecard model to make it more fair.
Bias Mitigation in Credit Scoring by Disparate Impact Removal (Risk Management Toolbox)
Use disparate impact removal as a pre-processing technique in bias mitigation to a credit scorecard model to reduce bias in the model.
Interpretability and Explainability for Credit Scoring (Risk Management Toolbox)
This example shows different techniques for interpreting and explaining the logic behind credit scoring predictions.
Credit Scorecard Modeling Workflow
Use the credit scorecard workflow to create, model, and analyze credit scorecards.
About Credit Scorecards
The goal of credit scoring is ranking borrowers by their credit worthiness.
Credit Scorecard Modeling Using Observation Weights
Use observation weights with the credit scorecard workflow to create, model, and analyze credit scorecards.

Troubleshooting

Troubleshooting Credit Scorecard Results

Troubleshoot results when using a creditscorecard model.