Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact

Repository Web            Google
View ALL Data Sets

× Check out the beta version of the new UCI Machine Learning Repository we are currently testing! Contact us if you have any issues, questions, or concerns. Click here to try out the new site.

South German Credit (UPDATE) Data Set
Download: Data Folder, Data Set Description

Abstract: 700 good and 300 bad credits with 20 predictor variables. Data from 1973 to 1975. Stratified sample from actual credits with bad credits heavily oversampled. A cost matrix can be used.

Data Set Characteristics:  


Number of Instances:




Attribute Characteristics:

Integer, Real

Number of Attributes:


Date Donated


Associated Tasks:

Classification, Regression, Clustering

Missing Values?


Number of Web Hits:



Ulrike Grömping
Beuth University of Applied Sciences Berlin
Website with contact information:

Data Set Information:

The widely used Statlog German credit data ([[Web Link](german+credit+data)]), as of November 2019, suffers from severe errors in the coding information and does not come with any background information. The 'South German Credit' data provide a correction and some background information, based on the Open Data LMU (2010) representation of the same data and several other German language resources.

Attribute Information:

## This section contains a brief description for each attribute.
## Details on attribute coding can be obtained from the accompanying R code for reading the data
## or the accompanying code table,
## as well as from Groemping (2019) (listed under 'Relevant Papers').

Column name: laufkont
Variable name: status
Content: status of the debtor's checking account with the bank (categorical)

Column name: laufzeit
Variable name: duration
Content: credit duration in months (quantitative)

Column name: moral
Variable name: credit_history
Content: history of compliance with previous or concurrent credit contracts (categorical)

Column name: verw
Variable name: purpose
Content: purpose for which the credit is needed (categorical)

Column name: hoehe
Variable name: amount
Content: credit amount in DM (quantitative; result of monotonic transformation; actual data and type of
transformation unknown)

Column name: sparkont
Variable name: savings
Content: debtor's savings (categorical)

Column name: beszeit
Variable name: employment_duration
Content: duration of debtor's employment with current employer (ordinal; discretized quantitative)

Column name: rate
Variable name: installment_rate
Content: credit installments as a percentage of debtor's disposable income (ordinal; discretized quantitative)

Column name: famges
Variable name: personal_status_sex
Content: combined information on sex and marital status; categorical; sex cannot be recovered from the
variable, because male singles and female non-singles are coded with the same code (2); female widows cannot
be easily classified, because the code table does not list them in any of the female categories

Column name: buerge
Variable name: other_debtors
Content: Is there another debtor or a guarantor for the credit? (categorical)

Column name: wohnzeit
Variable name: present_residence
Content: length of time (in years) the debtor lives in the present residence (ordinal; discretized quantitative)

Column name: verm
Variable name: property
Content: the debtor's most valuable property, i.e. the highest possible code is used. Code 2 is used, if codes 3
or 4 are not applicable and there is a car or any other relevant property that does not fall under variable
sparkont. (ordinal)

Column name: alter
Variable name: age
Content: age in years (quantitative)

Column name: weitkred
Variable name: other_installment_plans
Content: installment plans from providers other than the credit-giving bank (categorical)

Column name: wohn
Variable name: housing
Content: type of housing the debtor lives in (categorical)

Column name: bishkred
Variable name: number_credits
Content: number of credits including the current one the debtor has (or had) at this bank (ordinal, discretized
quantitative); contrary to Fahrmeir and Hamerle’s (1984) statement, the original data values are not available.

Column name: beruf
Variable name: job
Content: quality of debtor's job (ordinal)

Column name: pers
Variable name: people_liable
Content: number of persons who financially depend on the debtor (i.e., are entitled to maintenance) (binary,
discretized quantitative)

Column name: telef
Variable name: telephone
Content: Is there a telephone landline registered on the debtor's name? (binary; remember that the data are
from the 1970s)

Column name: gastarb
Variable name: foreign_worker
Content: Is the debtor a foreign worker? (binary)

Column name: kredit
Variable name: credit_risk
Content: Has the credit contract been complied with (good) or not (bad) ? (binary)

Relevant Papers:

Fahrmeir, L. and Hamerle, A. (1981, in German). Kategoriale Regression in der betrieblichen Planung. *Zeitschrift für Operations Research* **25**, B63-B78.

Fahrmeir, L. and Hamerle, A. (1984, in German). *Multivariate Statistische Verfahren* (1st ed., Ch.8 and Appendix C). De Gruyter, Berlin.

Grömping, U. (2019). South German Credit Data: Correcting a Widely Used Data Set. Report 4/2019, Reports in Mathematics, Physics and Chemistry, Department II, Beuth University of Applied Sciences Berlin. URL: [[Web Link]].

Häußler, W.M. (1979, in German). Empirische Ergebnisse zu Diskriminationsverfahren bei Kreditscoringsystemen. *Zeitschrift für Operations Research* **23**, B191-B210.

Hofmann, H.J. (1990, in German). Die Anwendung des CART-Verfahrens zur statistischen Bonitätsanalyse von Konsumentenkrediten. *Zeitschrift für Betriebswirtschaft* **60**, 941-962.

Open data LMU (2010; accessed Nov 27 2019; in German). Kreditscoring zur Klassifikation von Kreditnehmern. URL: [[Web Link]].

Citation Request:

Grömping, U. (2019). South German Credit Data: Correcting a Widely Used Data Set. Report 4/2019, Reports in Mathematics, Physics and Chemistry, Department II, Beuth University of Applied Sciences Berlin.

Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML