Connect with us


Backtesting of margin models




Given the recent increases in initial margin that I covered in a few recent blogs (see Procyclical margins in the time of Covid-19 and Crashing rates and swap margins), I wanted to look into backtesting of margin models.


Backtesting is a well established practice, widely used by all CCPs to check the adequacy of margin cover and by Banks for adequacy of capital required for market risk.

In a nutshell it checks the daily VM (or PL) observed over time for an account is less than the Initial Margin held for that account and quantifies the number of times that it is higher.

The output of a backtesting run looks something like this.

The squiggly line is the PL, in our case daily, over a time period from Oct 2016 to Sep 2018, while the horizontal line is the Initial margin, in this case $4.7 million.

There is one date, Jun 18, 2018 where the PL of $6.7 million exceeds the initial margin.

Margin period of risk (MPOR) refers to the number of days of market risk that initial margin is calibrated too and the PL should be over the same period. Typically CCPs use 1 or 2 days for Futures and 5 days for Swaps, while Uncleared Margin Rules (<a href="" title="Glossary: UMR" data-cmtooltip="Uncleared Margin Rules.  Refers to rules pertaining to collateralizing bilateral (uncleared) swaps, with both Initial and Variation margin.  The rules are implemented in various jurisdictions, beginning with the US and Japan in September 2016. This blog includes more details.” class=”glossaryLink” target=”_blank”>UMR) require 10-days.

Evaluating a Backtest

From the backtesting results, we need a few simple measures that tell us something about the soundness of the margin model:

  • the number of exceptions
  • the number of observations
  • the amount that the largest exception is in excess of margin

In our example, these are 1, 250 and $2 million.

Then given our design of the margin model and in particular the confidence level, we could assess if the model was adequate and passes the backtest or not.

Foe example for a 99% confidence level value-at-risk (VaR) model, we expect on average 1 out of 100 days to have an exception, so the above backtest result would pass, while 4 or more exceptions would cause as to doubt the adequacy of the model.

Note that using VaR as the metric for margin does not state anything about how large the exception could be beyond VaR, it could be 110% of VaR or 300% of VAR! This is where tail measures like Expected Shortfall (ES) are better as they account for the long tails we see in financial time-series data.

Basel II & III for Banks requires the use of a statistical traffic light test with green, amber & red zones for backtesting results, details of which can be found in the BCBS document “Minimum capital requirements for market risk” (January 2019).

CCP Quantitative Disclosures

For CCPs, we can turn to Quantitative Disclosures in CCPView for insights into backtesting results (Disclosures 6.5.n). Let’s do so for a few selected clearing services.

6.5 Disclosures from <a href="" title="Glossary: CCPView" data-cmtooltip="A Clarus service that aggregates, normalizes, and presents Clearing data on a weekly basis. Found at” class=”glossaryLink” target=”_blank”>CCPView

  • Eurex Clearing Fixed Income, shows 316 exceptions over the past 12 months on any account, out of a total of 219,454 observations, a coverage level of 99.86% and a peak uncovered exposure of $373 million
  • LCH SwapClear, shows 2,163 exceptions, 1,995,771 observations, a coverage level of 99.89% and a peak uncovered exposure of $157 million.

Note: Coverage level is defined as 1 – ( expressed as a percentage, where 6.5.2 is the number of observations and the number of times cover fell below mark to market exposure.

As the number of accounts are very different for each clearing service (e.g. LCH SwapClear has >23,000 client accounts) and the exceptions are over all the accounts, comparing the absolute exceptions between CCPs needs to be done carefully.

However the coverage level and size of uncovered exposure are directly comparable and show less variance between CCPs, which is what we would expect and be concerned by CCP results that were outliers.

It will be very interesting to see how these numbers change for 31-Mar-20, when we get these disclosures in early July.

I am sure CCPs themselves have a lot more backtesting results data and share some of that with members on their risk committees, possibly including the Green, Amber, Red Zone type tests.

Margin Models

In addition disclosures 6.4.n provide details to compare IM models, lets do so for the same four clearing services above.

6.4 Disclosures from CCPView
  • Three of these use Historical Simulation and one Monte-carlo Simulation
  • Confidence levels used are 99% or 99.7%
  • Look back periods vary from 500 days to 10 years, with Eurex noting the use of Stress periods
  • Close out period (MPOR) is either 1-day, 2-days, 5-days or 7-days.
  • No consistent disclosure on if VaR or ES is used, though OCC mention ES in 6.4.1

I am sure all these models are well calibrated and perform well in backtesting.

The real value in the disclosures is comparing across many CCPs for differences or outliers and questioning these with the CCPS risk committees, as well as noting changes over time.

The End

That’s all for today.

Backtesting is important for margin models.

It provides the confidence that initial margin is adequate.

The coverage level (exceptions/observations) is important.

So is largest uncovered exposure.

CCPView has disclosures for 36 Clearing Houses with many more CCPs.

Covering Backtesting, IM models and more.

Stay informed with our FREE newsletter, subscribe here.



Aite survey: Financial institutions will invest more to automate loan process




Financial institutions plan to increase their spend on automations and collections management solutions for their loan processes. Fresh results on consumer lending practice from research and advisory firm Aite Group indicate lenders plan to invest more heavily in their collections processes, said Leslie Parrish, senior analyst for the Aite Group’s consumer lending practice. Parrish shared […]

Coinsmart. Beste Bitcoin-Börse in Europa

Continue Reading


Facial recognition, other ‘risky’ AI set for constraints in EU




Facial recognition and other high-risk artificial intelligence applications will face strict constraints under new rules unveiled by the European Union that threaten hefty fines for companies that don’t comply.

The European Commission, the bloc’s executive body, proposed measures on Wednesday that would ban certain AI applications in the EU, including those that exploit vulnerable groups, deploy subliminal techniques or score people’s social behavior.

The use of facial recognition and other real-time remote biometric identification systems by law enforcement would also be prohibited, unless used to prevent a terror attack, find missing children or tackle other public security emergencies.

Facial recognition is a particularly controversial form of AI. Civil liberties groups warn of the dangers of discrimination or mistaken identities when law enforcement uses the technology, which sometimes misidentifies women and people with darker skin tones. Digital rights group EDRI has warned against loopholes for public security exceptions use of the technology.

Other high-risk applications that could endanger people’s safety or legal status—such as self-driving cars, employment or asylum decisions — would have to undergo checks of their systems before deployment and face other strict obligations.

The measures are the latest attempt by the bloc to leverage the power of its vast, developed market to set global standards that companies around the world are forced to follow, much like with its General Data Protection Regulation.

The U.S. and China are home to the biggest commercial AI companies — Google and Microsoft Corp., Beijing-based Baidu, and Shenzhen-based Tencent — but if they want to sell to Europe’s consumers or businesses, they may be forced to overhaul operations.

Key Points:

  • Fines of 6% of revenue are foreseen for companies that don’t comply with bans or data requirements
  • Smaller fines are foreseen for companies that don’t comply with other requirements spelled out in the new rules
  • Legislation applies both to developers and users of high-risk AI systems
  • Providers of risky AI must subject it to a conformity assessment before deployment
  • Other obligations for high-risk AI includes use of high quality datasets, ensuring traceability of results, and human oversight to minimize risk
  • The criteria for ‘high-risk’ applications includes intended purpose, the number of potentially affected people, and the irreversibility of harm
  • AI applications with minimal risk such as AI-enabled video games or spam filters are not subject to the new rules
  • National market surveillance authorities will enforce the new rules
  • EU to establish European board of regulators to ensure harmonized enforcement of regulation across Europe
  • Rules would still need approval by the European Parliament and the bloc’s member states before becoming law, a process that can take years

—Natalia Drozdiak (Bloomberg Mercury)

Coinsmart. Beste Bitcoin-Börse in Europa

Continue Reading


What banks are looking for after COVID-19




As the number of vaccinated adults in the United States grows and the average rates of COVID-19 infections drops, bankers and their customers are eager to move ahead. But what does “move ahead” look like, and how has consumer demand changed over the course of the past year?

To answer this question, we at MX surveyed 1,000 randomly selected U.S. consumers with results published in our ultimate guides. Among many insights, we found that 87% of consumers say they now visit their bank branch less often than they did before the COVID-19 pandemic, while 89% say they use mobile banking more often. We also found that a quarter of respondents said they’d currently feel insecure about their financial situation if they hadn’t received a third stimulus check, while half said that without the stimulus check they’d struggle to cover their monthly living expenses such as rent, mortgage, and recurring bills.

In other words, consumer demand for digital banking has surged at the same time that consumers have received an influx of stimulus cash — a shift that puts banks in a bit of a bind. As covered in an article from Bloomberg Opinion columnist Brian Chappatta, banks now face the largest gap in decades between a typical bank’s deposits at hand (which are high) and the demand for new loans (which is low). Given this gap, Chappatta writes, “In the near future, [banks] may need to rely even more heavily on revenue outside of their traditional business of making loans.”

Banks looking for these new revenue models in the wake of COVID-19 should know that the way forward must be digital first.

One possibility worth exploring is a subscription service, possibly in the vein of Amazon Prime. As Bradley Leimer, co-founder of Unconventional Ventures, says, “If banks can’t offer something more valuable than Amazon Prime, then we’re probably in the wrong business. I think we just need to retool our mindsets and put the customer at the heart of these decisions. What is at stake, in my opinion, is literally the future of the financial services model. The wolves are at the door.”

In light of this, financial institutions can ask themselves which benefits they offer could be packaged together for a monthly subscription. For example, Utah Community Credit Union (UCCU) has developed a product called UCCU Prime, which gives members services including $600 per claim in cell phone protection in the event their phone is broken or stolen, $80 in coverage for roadside assistance (4x a year), a $10,000 reimbursement in expenses in the event of stolen identity, $10,000 travel accidental death coverage, special deals for local businesses, travel discounts nationwide, debit card rewards, and more — all for $6 a month. As UCCU creates new offerings, they can add them to the bundle and increase this revenue stream.

This is just one of many possibilities that comes with looking at banking with a new pair of eyes as we work to put COVID-19 behind us and explore new revenue streams.

Learn more at

Coinsmart. Beste Bitcoin-Börse in Europa

Continue Reading


The Impact of COVID-19 on Capital Markets Operations




The Depository Trust & Clearing Corporation (DTCC), the premier market infrastructure for the global financial services industry, today published a white paper examining how capital markets operations responded during the COVID-19 pandemic and where market participants are focused in a post-pandemic future.

In a new white paper, “Managing through a Pandemic: The Impact of COVID-19 on Capital Markets Operations”, buy and sell-side firms reported that, while their post-trade operations (Ops) and operations technology (OpsTech) proved largely resilient during the pandemic, several key challenges emerged as market volatility surged throughout 2020. The study was conducted with research assistance from McKinsey & Company, and was based on insights from Ops and OpsTech professionals at 35 buy and sell-side firms. The top areas of focus highlighted were:

  • Cash fixed income and cash equities were most impacted by the pandemic-induced market volatility, with 30-35% of firms across the buy-side and the sell-side reporting operational post-trade processing challenges in these asset classes.
  • From a processing perspective, settlements/payments and collateral/valuations were impacted the most, with 58% of sell-side firms reporting challenges in settlement and payments during the peak of the pandemic.
  • Buy-side firms typically experienced less disruption to post-trade processes than sell-side firms due to simpler operational models, with the sell-side reconciling breaks and settling trades across hundreds of counterparties.
  • The sudden transition to a work-from-home operating model was achieved almost seamlessly, and in most cases within a matter of days, due to the ability to implement tactical changes to operating models.

Respondents cited that efforts made in recent years to re-engineer and automate processes and upgrade technology platforms were the main reason for firms’ resilience during the pandemic and their ability to manage an unusually prolonged business-continuity planning (BCP) event. A significant majority of respondents stated that the pandemic validated their Ops priorities and investment plans.

Michael Bodson, President & CEO at DTCC, said, “During, and in the immediate aftermath of the COVID-19 pandemic, the industry remained resilient, with buy and sell-side firms working seamlessly to support unprecedented volumes and ensure uninterrupted trading for clients and underlying investors. However, opportunities remain for further optimizing post-trade processes across the capital markets.”

The survey highlighted that while the pandemic did not create an impetus for change in Ops and OpsTech due to largely resilient operations, a consensus emerged around where firms should focus next:

  • Sell-side and buy-side firms are aligned on the need to further simplify and standardize a sub-set of post-trade services which were hardest hit. For the sell-side, these include making enhancements to reconciliations and confirmations capabilities, while the buy-side prioritized an increased focus on fails and collateral management.
  • More than half of firms who responded to the survey plan to increase capacity, build new capabilities or re-engineer post-trade processes.
  • Respondents highlighted the need for a continued focus on shortening settlement cycles due to the impact of the unprecedented trading volumes and volatility on liquidity and margin. More than 50% of firms plan to increase capacity in support of these processes.
  • The stigma around working from home and productivity no longer exists, with many firms planning to retain part of the remote and flexible working model post-pandemic across post-trade operations.

Bodson added, “As the impact of the pandemic continues to unfold, firms must keep their focus on delivering continued improvements to efficiency, while reducing risk. At the same time, to unlock new sources of value and remain relevant to clients, a focus on innovation will be essential. The industry will need to embrace collaborative approaches, common processes, best practices and deploy operating models that continue to meet the evolving needs of market participants.”

Coinsmart. Beste Bitcoin-Börse in Europa

Continue Reading
Esports5 days ago

Dota 2 patch 7.29b brings nerfs to Phantom Lancer and Lifestealer amongst other hero balance changes

Esports5 days ago

Apex Legends Season 9 will add new hero, fix Banglore bugs

Esports5 days ago

Code S: Trap & Zest advance to RO8, playoff bracket set

Esports5 days ago

New CSGO Update Makes Items Purchased From Store Non Tradable for a Week

Blockchain3 days ago

Mining Bitcoin: How to Mine Bitcoin

Esports5 days ago

Radiant Valorant streamer Solista banned for cheating on live stream

Esports5 days ago

How to Calculate Steam Market Tax on CSGO Items

Fintech4 days ago

Fintech offers brokers better commissions after BID

Esports4 days ago

OWL 2021 Power Rankings – #9 Guangzhou Charge

Esports5 days ago

OWL 2021 Power Rankings – #10 Washington Justice

Blockchain4 days ago

Stanislovas Tomas im Interview: „NFTs können unsere Gesellschaft verändern“

PR Newswire3 days ago

Hello Pal Signs Definitive Purchase Agreement to Acquire Interest in Dogecoin/Litecoin Mining Assets

Esports4 days ago

xQc Banned From NoPixel GTA RP Server Once Again

Esports4 days ago

CDL Challengers Elite Stage 3 Preview

Coinbase hourly chart
Blockchain3 days ago

Coinbase Addresses Future Revenue Concerns With Plans to Become Crypto’s Amazon

Esports5 days ago

Zayt Retires From Competitive Fortnite For The Second Time

Esports4 days ago

Twitch streamer Lando Norris takes Italian F1 Grand Prix podium

Esports5 days ago

Dota 2: DPC Weekly Recap — SEA Apr 12-17, 2021

Esports4 days ago

Three takeaways from the SWT Japan Ultimate Online Qualifier

Esports4 days ago

Cloud9 Perkz says Kassadin can’t ever be balanced in LoL