german.credit: German Credit Data

In fairml: Fair Models in Machine Learning

German Credit Data

Description

A credit scoring data set that can be used to predict defaults on consumer loans in the German market.

Usage

data(german.credit)

Format

The data contains 1000 observations (700 good loans, 300 bad loans) and the following variables:

• Account_status: a factor with four levels representing the amount of money in the account or "no chcking account".

• Duration: a continuous variable, the duration in months.

• Credit_history: a factor with five levels representing possible credit history backgrounds.

• Purpose: a factor with ten levels representing possible reasons for taking out a loan.

• Credit_amount: a continuous variable.

• Savings_bonds: a factor with five levels representing amount of money available in savings and bonds or "unknown / no savings account".

• Present_employment_since: a factor with five levels representing the length of tenure in the current employment or "unemployed".

• Installment_rate: a continuous variable, the installment rate in percentage of disposable income.

• Other_debtors_guarantors: a factor with levels "none", "co-applicant" and "guarantor".

• Resident_since: a continuous variable, number of years in the current residence.

• Property: a factor with four levels describing the type of property to be bought or "unknown / no property".

• Age: a continuous variable, the age in years.

• Other_installment_plans: a factor with levels "bank", "none" and "stores".

• Housing: a factor with levels "rent", "own" and "for free".

• Existing_credits: a continuous variable, the number of existing credit lines at this bank.

• Job: a factor with four levels for different job descriptions.

• People_maintenance_for: a continuous variable, the number of people being liable to provide maintenance for.

• Telephone: a factor with levels "none" and "yes".

• Foreign_worker: a factor with levels "no" and "yes".

• Credit_risk: a factor with levels "BAD" and "GOOD".

• Gender: a factor with levels "Male" and "Female".

Note

The variable "Personal status and sex" in the original data has been transformed into Gender by dropping the personal status information.

References

UCI Machine Learning Repository:
https://archive.ics.uci.edu/ml/datasets/statlog+(german+credit+data)

fairml documentation built on May 31, 2023, 6:02 p.m.