24 October 2014

44975

Retired Member

644 | posts 1,993,570 | views 796 | comments

Financial Risk Management

This network brings together professionals involved in the oversight and management of their company's financial risks and exposures as well as solution vendors, in order to discuss risk issues including interest rate risk, foreign exchange risk and commodity price risk, among others.

Scorecards. Achieving Most Precise Borrower Segmentation

19 November 2012  |  2323 views  |  0

Below we examine most common mistakes in scorecard development, and explain how to use them for most precise borrower rating and segmentation.

Scorecards provide a set of weights assigned to characteristics that demonstrate customer’s credit worthiness. With scorecards, customers are evaluated and rated according to their potential probability.

You can automatically make profitable decisions once customers are evaluated and scored: approve the loan application, define optimal pricing, take advantage of the up-sell opportunitites etc.

I'd recommend reading The Very Basics of Scorecards by Brendan Le Grange in his Credit Risk Strategy Blog, if you're new to the subject: this is a clear and elegant article on the subject.

Mistakes in scorecard development can be made in each of the following stages – data sampling, statistical evaluation of borrowers’ characteristics, formation of training and validation datasets.

1. Data sampling. Inaccuracies during data sampling are caused by selecting data samples that are not responding to the requirement of representativeness and randomness.

Representativeness means the maximum proximity of the data indicators in the sample to the actual borrower's characteristics in the loan portfolio. This requirement is natural and understandable, because the scorecard is expected to reflect specifics of the dataset used for its development.

Randomness implies that loan application data should be included in the working sample independently.

There are two ways to prevent sampling mistakes – either by directly controlling data sampling procedures, or by evaluating the borrower’s statistical characteristics.

2. Statistical evaluation of borrowers’ characteristics. When evaluating borrower’s statistical characteristics you should pay attention to those indicators that are unnaturally distributed. For instance:

  • Indicators that characterize one prevailing borrower category;
  • Characteristics pointing to an obvious gap between selected groups of borrowers and actual characteristics presented in the credit portfolio.

Once you have discovered unequally or unnaturally distributed characteristics, you should change or adjust the procedure of forming the working sample and set different rules for assigning values to each characteristic.

3. Selection of training and validation datasets. The main criteria for selecting a training data sample says: training data should provide enough examples of profitable and delinquent loans.

As a rule, a training dataset comprised of 3500-4000 records is enough to successfully train a scorecard; provided that this dataset offers a 3:1 proportion of “good” and “bad” loan cases. It is possible to train a scorecard with fewer records, but you should keep the 3:1 proportion of the loan case profitability.

This way, data mining techniques can be applied to improve borrower rating and segmentation in automated loan application processing. In upcoming articles, we will show more ways to apply data mining for decision automation.

TagsRisk & regulationRetail banking

Comments: (0)

Comment on this story (membership required)
Log in to receive notifications when someone posts a comment

Latest posts from Retired

7-day account switch: customer empowerment or indifference

18 June 2014  |  1840 views  |  1  |  Recommends 0 TagsRisk & regulationRetail banking

On Reinventing Money.

03 June 2014  |  1237 views  |  0  |  Recommends 0 TagsPaymentsInnovation

Operational Lessons for New Boutique Asset Managers

27 May 2014  |  623 views  |  0  |  Recommends 0 TagsRisk & regulation

Trading System Failures Cannot Be Our Norm

21 May 2014  |  1479 views  |  0  |  Recommends 0 TagsRisk & regulationInnovation

Around the Clock Tweeting

15 May 2014  |  1946 views  |  0  |  Recommends 0 TagsMobile & onlineRetail banking
name

Retired Member

job title

company name

member since

2014

location

Summary profile See full profile »

Retired's expertise

What Retired reads
Retired writes about

Who is commenting on Retired's posts

Rasvan Stanescu
Sian Bentley
Tony Wenzel
Jorge Yui
Ketharaman Swaminathan
Mark Pavan
Vishal Chaturvedi
Matt Scott
Geoffrey Barraclough
Thad Peterson
Marinka Ryan
Alexander Peschkoff