Post Trade Insights

Accelerate ISDA SIMM calculations with algorithmic differentiation 

Post Trade Solutions

In today’s competitive and fast-paced financial landscape, efficiency and accuracy in risk management are crucial. When it comes to derivatives margin reconciliation and related exchange workflows, ISDA Standard Initial Margin Model (SIMM™) calculations (and the underlying delta and vega sensitivity inputs) can be a lengthy and resource-intensive process. 

This is particularly true for financial institutions dealing with larger portfolios containing complex products such as Best-of/Worst-of Basket Swaps, Knock-in/Knock-out (KIKO) Variance Swaps and Corridor Variance Swaps. Calculating the thousands of underlying trade sensitivities for these products can take several hours if run in a conventional bump and revaluation approach – creating operational bottlenecks and challenges when re-runs are needed to incorporate new trades or fix errors. 

To tackle this long-standing challenge, Post Trade Solutions has implemented a major enhancement to its Initial Margin Risk Generator, the service for automating initial margin calculations. By leveraging Algorithmic Differentiation – a mathematical solution for calculating derivatives for efficient use of the chain rule – the update significantly reduces computational run-time, making the sensitivity calculations much faster and more scalable for complex portfolios.

Matching speed with accuracy 

Algorithmic Differentiation (AD) provides a cost-effective alternative to expensive hardware such as graphics processing unit (GPUs) and/or multiple central processing unit (CPU) or cloud-based calculation cores, which can be costly and quickly become outdated. It also maintains accuracy while improving speed and operational efficiency, unlike many other software-based “short-cuts” or approximations. 

Post Trade Solutions is able to leverage AD through its innovative Scripted Trade framework, which allows clients to evaluate complex derivative payoffs under a common scripting language, and value the trade within a Monte Carlo simulation framework. The framework can be parameterised under a multivariate Black Scholes or Gaussian process and contains support for additional high-performance computing enhancements beyond AD – such as parallelisation.

Since the required trade sensitivities for ISDA SIMM™ are partial derivatives of the price with respect to each underlying risk factor, AD exploits simple arithmetic functions that have known analytical derivatives to calculate sensitivities more quickly than the alternative “bump and revalue” approach. “Bump and revalue” requires one additional pricing step for every underlying risk factor, which can number into the thousands for even just one complex derivative. 

The benefits of Post Trade Solution’s implementation are already being realised. The first client to implement the update experienced a 90-95% reduction in runtime, bringing calculations down from over two hours to mere minutes.

By offering the combination of speed and accuracy, this transformation empowers firms to reallocate their time and valuable resources to other areas where they’re needed most.

Looking ahead

The enhancement is currently being rolled out to additional clients, with full deployment taking place in April 2025. The AD implementation is also made freely available for the industry to explore at www.opensourcerisk.org under Example 61.

As the derivatives industry evolves, LSEG remains focused on delivering smarter, faster, and more innovative tools to help our clients navigate complexity and optimise performance. 

Legal Disclaimer

Republication or redistribution of LSE Group content is prohibited without our prior written consent. 

The content of this publication is for informational purposes only and has no legal effect, does not form part of any contract, does not, and does not seek to constitute advice of any nature and no reliance should be placed upon statements contained herein. Whilst reasonable efforts have been taken to ensure that the contents of this publication are accurate and reliable, LSE Group does not guarantee that this document is free from errors or omissions; therefore, you may not rely upon the content of this document under any circumstances and you should seek your own independent legal, investment, tax and other advice. Neither We nor our affiliates shall be liable for any errors, inaccuracies or delays in the publication or any other content, or for any actions taken by you in reliance thereon.

Copyright © 2025 London Stock Exchange Group. All rights reserved.