Credit Score Factor

From Open Risk Manual


Credit Score Factors are indicators, attributes, characteristics that can be used (usually after being converted to a numerical form) as ( inputs) in the construction of a numerical Credit Score, a procedure that is termed a Scorecard. Credit score factors constitute Explanatory Variables, that is, they aim to explain risk outcomes

Factor Types

Credit score factors can be categorized in various ways (inheriting from Risk Factor):

  • Content: Quantitative versus Qualitative Risk Factors, indicating the degree to which factors can be associated with concrete data
  • Temporal Characteristics: Eg., Static versus Dynamic factors, that is factors that change in time (with different possible timescales)
  • Variable Type (Numerical Categorical)

By Content

  • Quantitative Variables encode credit factors that are easily quantifiable (e.g., Financial Statements)
  • Qualitative Variables encode credit factors that capture a qualitative aspect that is not numerical in nature

By Temporal Properties

  • Static factors do not change in time (at least for the observation window of interest) e.g. nationality
  • Dynamic factors change either because of economic or other volatile phenomena. Dynamic timescales can be fast or slow (in relation with the Risk Horizon being considered. In the latter case we speak of Secular Evolution

By Variable Type

  • Numerical Variable. Numerical variables can be further of either
    • Integer Type, or
    • Real Type
  • Categorical Variable. Categorical variables range along a finite number of possible values (with the case of binary outcomes being sometimes denoted a Dummy Variable). They can be further distinguished as
    • Ordinal Variable, where the values satisfy a clear ordering relation (can be sorted)
    • Generic categorical, where the values do not form any ordered set

It is quite common to convert numerical variables into ordinal variables using binning, creating what is denoted a Binned Variable

Examples of Credit Score Factors

The following is an indicative list of factors and how they might be affecting the credit score:

  • Having established Credit History. Not having history adds uncertainty
  • Duration of Credit History. The longer the credit history the more the comfort that there is ability and willingness to repay
  • Type of borrowing (mortgage, credit card, auto loan, student loan etc). The presence of other borrowing indicates a positive evaluation from other lenders
  • Total Credit Limit. The size of the credit limit indicates that confidence of other lenders
  • Total credit balance. A high balance may indicate over-extended economics
  • Credit event history (Delinquency, Bankruptcy, Foreclosure, Repossession)

Issues and Challenges

  • There is sometimes confusion between qualitative factors and categorical variables. While qualitative factors are represented as categorical variables (if at all), quantitative factors can be represented as numerical variables or as categorical variables (via various form of discretization, binning etc.)
  • The conflation between correlation and causation that is common in many areas of Quantitative Risk Management
  • The ability of the totality of identified credit risk factors to explain the risk being considered (Unexplained Variance))
  • The influence of different risk factors on outcomes may be overlapping and difficult to disentangle (correlation / dependency between factors), in the extreme case, Collinearity
  • The impact of a credit risk factor on the credit risk profile may not be monotonic (See Non-Monotonic Risk Factor)
  • Certain credit risk factors may be very difficult to quantify / model. Qualitative factors are in general those that are difficult to quantify due to complexity, or Emerging Risk nature

See Also


This entry annotates a FIBO Ontology Class. FIBO is a trademark and the FIBO Ontology is copyright of the EDM Council, released under the MIT Open Source License. There is no guarantee that the content of this page will remain aligned with or correctly interprets the concepts covered by the FIBO ontology.