A high profile Australian virtual bank with a presence in NSW, VIC, QLD, WA and SA, was reviewing its monitoring and managing of liquidity risk.
Our client needed to develop and implement a solution to comply with the Basel reforms, and the Australian Prudential Standards (APS210). To meet these standards, our client needed to undertake scenario analysis on a regular basis and maintain minimum liquidity holdings. This required the development of a model capable of completing the following:
- Demonstrate that it has sufficient liquid holdings for a 30-day period under severe stress scenarios
- Demonstrate that it can continue to operate for at least 5 business days under adverse operating circumstances specific to it
- Model expected behaviour of cash flows over 15 months, or under the going concern scenario
A solution was needed that could handle the data intensity, link to numerous disparate business systems, and provide a robust and efficient daily process.
Parity developed an LCR model utilising the Microsoft Power BI suite of tools. This provided several advantages, including:
- A flexible, desktop solution
- The ability to collect and consolidate detailed account level data from a variety of sources
- Computational speed
Parity collaborated closely with the client to understand and interpret APS210 requirements, converting these into logical rules and a data model. Parity also provided training and support during the validation and acceptance phases of the project, as well as guidance on the IT and systems integration.
The completed model provided the client with insight to its risk-adjusted liability profile. It was run either daily or in real time as required. In addition to meeting regulatory requirements, the model provided the client with the insight and analytics to run a liability structure that is optimised for risk and its specific business.
“Parity collaborated closely with the client to understand and interpret APS210 requirements, converting these into logical rules and a data model.”
As a natural extension, the model also facilitates customer level analytics, insight into deposit flows, price sensitivity and customer behaviour. The model facilitates this though a cube-type functionality that allows data drill-down for both standardised reporting and ad hoc analysis.
Brownlow Medal Prediction Using Excel Based Machine LearningShare this articlelinkedintwitterA spreadsheet-based regression model using Neural Networks to rank this year’s medal contendersBy Stephen Huppert and Jack LanghammerIt may not be September, and the finals...
Our client is a leading Australian independent school that regularly places in the top 10 schools for academic results in Victoria. The client needed various…
The trustee to one of the leading industry super funds in Australia was evaluating alternative strategies to reduce costs to its members, including the option…