Get the latest updates and news from FINCAD. Subscribe and never miss a post! 


The 3 Keys to Future-Proof Curve Building
By Tony Webb, PhD | November 17, 2015

As most of us well know, accurate curve building is essential to the success of any derivatives business. But in recent times, erratic market behavior has forced firms to rethink their approaches to curve building. With this issue in mind, I recently conducted a webinar titled Flexible Curve Building: Valuation and Risk for Today and Tomorrow. In it, I explored common challenges organizations face around curve building and gave tips for overcoming these issues using an advanced and flexible curve building framework.

Over the last few years, we’ve seen rapid and, at times, volatile change happening in the capital markets. One noteworthy manifestation of this change has been the widening of interest rate basis spreads. The credit crisis of 08-09 saw many basis spreads blow up including, for example, the OIS-Libor basis. You may recall that at one point in 2008, the 3m USD Libor - Fed Funds spread shot up to over 400 basis points (bps) before settling back to just around tens or twenties of bps. This unforeseen event shattered the idea that Libor was a risk free rate.

Another change we’ve seen is greater regulation around margining. This has included the need to clear through central counterparties, and upcoming requirements to post margin for non-centrally cleared derivatives. These changes impact discount curves because of the funding rate implied by the collateral (e.g., OIS).

Yet a third example of change in the markets is the emergence of negative interest rates, which have been increasingly adopted in a rather unconventional effort to revive the economy. The dip into negative rates territory has presented some challenges for financial organizations, because many of their systems have been designed to assume non-negative rates.

What all of this uncertainty means is that it is becoming more necessary for firms to adopt a curve-building approach that is adaptable and resilient in the face of change. In order to accomplish this goal, firms need to ensure that their curve-building framework has the following key characteristics:

1. Use of Generic Concepts

Using generic concepts in your framework simplifies the curve building problem by breaking it down into pieces of a puzzle. Each part of the puzzle can then be solved separately. Additionally, the pieces can be swapped out with minimal effort, and without impacting other parts of the solution. They can also be reused in combination with other different pieces to solve new problems. This is the approach that FINCAD has taken in its curve building framework, where the concepts include “Target Method”, “Source Method”, “Calibration Target”, “Solver/Optimizer”, “Market Instrument”, etc.

Using generic concepts makes it easier to investigate the impact of different aspects of the curve building methodology, and ensures that the choices you ultimately make are identified explicitly. This approach also makes it quicker and less expensive to adapt to changes in market practices, thereby leading to higher profitability.

2. Risk Enabled Framework

The quantification of risk is nearly always a core requirement in applications that use curves. As such, a robust curve building framework must be designed with risk in mind, providing the ability to calculate risk efficiently, both in terms of numerical results and performance.

To optimize your results, ensure your valuation framework utilizes analytic risk as opposed to bumping for performing calculations. And since curve building is at the heart of valuation, the same applies to your curve building framework. While many firms use the traditional method of finite difference or “bumping,” this approach to calculating sensitivities requires at least one revaluation per risk factor and therefore can be inordinately slow. As a result, teams with complex multi-asset, multi-currency portfolios that use this method will often cut corners to reduce run-time. Typically they will elect to not bump every quote, or bump curves using a parallel shift, twist, or other aggregate bump. They are forced to make judgments of which risk factors matter most.

However, using analytic risk in the form of Algorithmic Differentiation (AD) brings to light all sensitivities to your inputs, giving you remarkable performance improvements—especially when multiple risk factors are involved. Furthermore, results using AD can be hundreds or thousands times faster when compared to bumping. All of this leads to more accurate hedging and locking in of profits.

3. Consistency

A robust curve building framework should have consistency, in the sense of self-consistent modeling and arbitrage-free pricing. It should also be able to enforce consistency across the enterprise, helping you eliminate silos.

Ensuring that curves are used in a self-consistent way will help you reduce the incidence of expensive modeling mistakes and get a better handle on model risk. Additionally, a framework that gives you a consistent view of risk across the organization makes it easier to manage the process of sharing curves and their associated assumptions with all relevant stakeholders.

A good way to achieve this level of continuity is to use three important ideas in your analytics architecture. These are the product (the trade or portfolio), the data model (a collection of self-consistent modeling assumptions and calibrated curves) and the valuation method. By utilizing these three dimensions, it’s possible to have a system where trades, models and numerical methods can all be hot swapped. This will let you easily change your curve building approach, which is embedded in the data model, while keeping everything else the same.

So, using a sophisticated curve building framework that (i) uses generic concepts, (ii) calculates risk efficiently and accurately, and (iii) enforces consistency, will prevent the need for you to waste time and money reengineering systems each time the market shifts, helping you future-proof your firm. It will also help you retain an edge, as you will be able to respond to changes potentially faster than many of your competitors.

For more information on how an advanced pricing and risk solution can help you adopt a flexible curve building framework, please view our on-demand webinar: Flexible Curve Building: Valuation and Risk for Today and Tomorrow

About the author
Tony Webb PhD
Tony Webb, PhD
Quantitative Advisor | FINCAD

Tony Webb is an experienced manager and quantitative analyst at FINCAD having held various roles, including Director of Analytics, VP R&D, Product Manager, and manager of technical Pre-Sales Analysts in NYC. He holds an MA in Mathematics from Cambridge University, a PhD in computational fluid dynamics from UBC, and an MBA with a specialization in finance from UBC. He is currently acting as a Quantitative Advisor within the Client Services department.