Risk Modelling

Base behaviour includes Automation of Underwriting with Analytics assessment at a unit level of financial product origination say loan origination. Secondly, the behaviour includes assorted provisioning of financial product templates, (translated as an offer say a loan offer). Thirdly, the behaviour includes generating control methods and triggers for Risk Mitigation driven by systematic reporting, both inside and outside the system.

  1. Automation of Risk Modelling for a single origination application: A decision journey screens and filters eligible applications and associates an assessment score based on a calibrated model. The bulk of originated applications is either auto-approved or rejected while flagging a few for supervision.

2. Product mapping with the originated application to translate into a product offer: A matchmaking algorithm that depends on already configured products and the relationship between the scorecard of the stated origination application.

3. Identifying a library template of the variable configuration of a Product. Identification of static factors with dynamic values, e.g. loan product defined with ranges of amounts in consideration such as Loan Amount, Approved Loan Amount, Disbursal Loan AmountAbility to group these products for an entity to entity-based mapping. Such as loan products offered to a specific group e.g. in case of salary advances, Employers, or companies as entities.

Another example is women-oriented loans or income ranges.

The loan product association with configurable factors is enhanced by the data model below.

4. Reporting and information representational standards of such a machine-type of Analytics Reporting Standards.

5. Identification of control methods and recovery triggers for Risk management(both manual and auto)Loan Portfolio at Analytics is tied to Loan Portfolio Below PAR, Above PAR, and other such classifications. Factors such as Loan Loss provisioning further require establishing a sound relationship with control methods and their application. It is also required to identify and assess the role of machine learning models that supersede the manual rules intervention in a non-directional movement of data process.

6. The configurable module of Risk Management provides features for customizing different data models and entity-to-entity mappings.

Risk Management with Finscale follows a Simulation-based approach; Using a combination of Finscale micro-services such as Analytics, Customer, and Financial Products, you can;

Simulate Risk over test data or historical data without affecting system performance using Finscale Analytics;

System Administration: Dynamic Score Card Generator:

System Administration: Business Rules for the Lending use case, please go to Credit Underwriting & Risk Profile AssessmentSystem Administration: Business Rules for Savings & Checking including digital wallets;

System Administration: Business Rules for Insurances;

Last updated

© Muellners Foundation. All Rights Reserved