SME Credit Scorecard

From Open Risk Manual

Definition

An SME Credit Scorecard is a type of Credit Scorecard used in the classification (scoring) of Credit Risk for SME Lending. The scorecard output is an assessment of the relative likelihood of a certain Credit Event occurring, given a number of observable inputs

SME Scorecard Types

By usage

SME Credit scorecards may be used for a number of distinct activities

By type of output

A scorecard might produce as output a Credit Score or an actual estimated Probability of Default

By nature of development

Broadly speaking, SME credit scorecards can be divided into two development type categories

  • Quantitative Scorecards (Credit Scoring Models) that use exclusively or primarily quantitative inputs and algorithmic processing (Machine Learning) to achieve the risk classification
  • Expert Scorecards that are developed by human subject matter experts


In practice quantitative scorecards might include expert based adjustments (Overrides) that aim to incorporate additional information not captured by the scorecard but available to the decision maker

By algorithm

Quantitative scorecards employ one of several possible algorithmic classes (For complete list see Credit Scoring Models), e.g. linear discriminant analysis, logistic regression, decision trees etc. According to [1] there are four methodological forms of multivariate credit scoring models:

  1. the linear probability model
  2. the logit model
  3. the probit model, and
  4. the multiple discriminant analysis model

SME Scorecard Development

Development of SME scorecards is highly dependent on context. An overview is given in the How to Build an SME Credit Scorecard entry.

Issues and Challenges

  • The size of SME varies smoothly between micro-enterprises that are similar to Consumer Finance and medium sized enterprises that are similar to Corporate entities. The segmentation into collectively managed and individually managed entities (and corresponding scorecards and Credit Rating based methodologies) is somewhat arbitrary
  • Like with any statistical model, the predictive performance of scoring models exhibits Model Decay, as both the borrower's own characteristics may change and the external business and economic conditions evolve

See Also

References

  1. L.Allen, G.DeLong, A.Saunders, "Issues in the credit risk modeling of retail markets 2004, Journal of Banking & Finance