Risk management technology product of the year: Fincad

Bob Park

Pricing and valuing derivatives has become a thornier problem as the product spectrum has widened. The obvious solution, a library of product-specific analytics, faces a challenge of its own – namely, how to keep up with new modelling insights or structuring tweaks.

Some vendors have made their name by being nimble, while others have offered structuring tools with varying degrees of power. In 2006, Vancouver-based Fincad – which has offered an analytics library since the early 1990s – decided to rethink the issue.

“Libraries are focused on individual instruments, but we are now in a more portfolio-orientated world,” says Bob Park, the company’s co-founder, president and chief executive. Fincad also realised that by simply extending its existing library, it would be unable to benefit from advantages offered by new technology developments, such as modern object-oriented programming, which requires a different design approach.

The result was F3 – a generic system that can price almost any financial instrument, whether simple or complex, and measure its risk. Fincad spent four years and tens of millions of dollars developing F3, and it was a risky move – the market was accustomed to the library approach and Fincad had to wait for its alternative to be accepted after its launch in 2010. But the past year has seen it gain widespread recognition. The product now has 70 institutions using it, and Fincad was voted most innovative vendor in last year’s Risk technology rankings (Risk December 2012, pages 58-65).

“What we did with F3 was to decouple the financial engineering from the software engineering,” says Park. “By separating the description of trades and portfolios from the construction of the financial models, and the valuation method employed, F3 provides a generic way to represent any trade or financial structure.”

Libraries are focused on individual instruments, but we are now in a more portfolio-orientated world

In addition to this new architecture, F3 had to make a number of other innovations. A generic calibration engine enables users to calibrate all parameters in an F3 model to any relevant market data – potentially leading to big time savings as market data often has to be calibrated for each instrument before pricing. A patent-pending method Fincad calls universal risk technology automatically calculates all the first-order risk sensitivity values for every data point used in the construction and analysis of trades and portfolios. The sensitivities are calculated using fast analytical differentiation rather than the usual time-consuming ‘curve bumping’ method, in which yield curves are moved up and down and the trade is painstakingly revalued each time. F3 also records all calculations and edits executed by users, so valuation and risk analysis can be regenerated at any time – something that has become more important for regulatory compliance. And the software has interfaces to run in Excel or MatLab – or a software development kit (SDK) enables it to be integrated into the user’s IT infrastructure, and can run on computer grids for high performance and scalability.

Aberdeen Asset Management, with £187 billion in assets under management, was one of the first users of F3. The firm was looking for software to price over-the-counter derivatives, particularly interest rates, inflation and credit default swaps, which offered both an SDK and Excel interface – which not every vendor does – as well as a tool that would enable the firm to price instruments itself, rather than having to wait for the provider to update its library.

“F3 works from first principles – you tell it every aspect of the swap and the pricing model you wish to use. We wanted that complete flexibility,” says Matthew Lynes, portfolio manager at Aberdeen Asset Management. The firm is now using F3 across its global trading desks.

Valuation service providers and hedge funds are among other institutions that require flexibility with their analytics, not only because of the wide range of instruments they must cover but also for model transparency and accuracy. Pricing specialist DerivActiv, for example, uses F3 to check the accuracy of its in-house model. “F3 allows us to configure every aspect of the modelling. This means we can value instruments in many different ways. We need that ability in order to compare the accuracy of our models and troubleshoot them when we find problems,” says Orlando Calvo Arias, a Panama-based quantitative analyst at the firm.

Fincad is continuing to invest in the development of F3, says Park – for example, by giving users the ability to tune down the heavyweight analytics when they face simpler pricing problems. “Customers told us F3 could handle some of the gnarliest problems they had but was too difficult to use for more vanilla products, so we added short cuts and productivity tools to make the easy stuff more doable,” says Park.