Every so often, markets become dominated by so called “disruptive technology” solutions that offer a genuine order of magnitude improvement in functionality, speed, flexibility and cost. Cubes or OLAP (online analytical processing), which are the cornerstone of all good business intelligence platforms, are threatening traditional database approaches to risk management and reporting. Major financial institutions and commodities companies such as Barcap, Nomura, Centrica, EDFTrading and Gazprom, are adopting Cube technology to create a range of credit and market risk solutions that offer sparkling performance – No more error prone batch processes producing inflexible fixed reporting, as all risk results are computed on the fly. The ‘Cube’s’ real-time reporting capability delivers answers to any query in seconds. The ultimate in flexibility – A range of self-service reporting tools allow end users to access their own data, delivering the right information, at the right time in the right format. With simple drag and drop techniques, coupled with fast queries, a range of reports can be constructed from simple summary reports to complex data analysis.
Getting started with Cubes
Cubes are available from a large range of suppliers such as Microsoft, Oracle and IBM. Whilst I only have experience with Microsoft’s analysis services, there are a number of pre-requisites to any good Cube project. Expert domain knowledge; A good Cube will require a good degree of expert knowledge of the target domain. Only with this knowledge can a good star schema be developed that will provide a Cube design to fulfil the target users needs. (A star schema is essentially the equivalent in Cubes of a good database design). Design expertise in the chosen business intelligence tool – The key to the Cubes sparkling performance will be to utilise an expert in the chosen business intelligence tool. Time intelligence – The need to analyse data using time as a parameter, is a major component of good Cubes for risk. Regulators demand that a risk department can demonstrate what could happen in the future, based on certain factors, and Cubes provide an excellent platform to enable this sort of reporting. Compliance users typically want to look at what happened in the past, and spot any unusual behaviour or new trends. Again cubes are an excellent vehicle for this sort of analysis. Visualisation – There is a large range of visualisation tools available to query and report on Cube data. We have found that Excel is an essential visualisation tool in the risk management arena. Risk managers, if faced with fixed reporting formats, will inevitably find it hard to dispense with their favourite spread sheets. Make sure whatever tool you use has good Excel interoperability.
If you follow these simple guidelines it will lead to a rapid return on investment and a risk management solution that is resilient, flexible and fit for purpose for the next 20 years.