Pulling forward thanks to our knowledge.

Pulling forward thanks to our knowledge.

Please find below the latest news and publications.

News & Publications

June 2017

InCube lanciert SA-CCR Calculator

InCube Advisory hat mit dem InCube SA-CCR Calculator eine einfache, flexible und transparente Lösung zur Berechnung der Kreditäquivalente von Derivaten sowie der risikogewichteten Aktiva (RWA) auf Basis eines Excel-Workbooks entwickelt. Alle Berechnungen folgen dem neuen Standardansatz (SA-CCR) bzw. dem vereinfachten Standardansatz (VSA-CCR) und erfüllen damit die Vorgaben des FINMA Rundschreibens «Kreditrisiken Banken» (FINMA RS 17/7) sowie der Eigenmittelverordnung (ERV, 952.03).

Bereits mehr als 20 Banken haben sich für die Verwendung des InCube SA-CCR Calculators entschieden. Hier eine Übersicht ausgewählter Merkmale:

  • Die Verwendung von Zellenformeln sowie die transparente Darstellung sämtlicher Zwischenergebnisse gewährleisten jederzeit vollständige Nachvollziehbarkeit aller Berechnungen (Audit).

  • Informative Beschreibungen zu den Input-Daten und Berechnungsschritten sowie Verweise auf relevante Randziffern des Rundschreibens erlauben das einfache Nachschlagen relevanter Informationen.

  • Alle Daten können jederzeit manuell befüllt oder automatisiert aus .csv/.txt Dateien geladen werden. Dabei gewährleistet eine parametrisierte Datenschnittstelle die einfache Anpassbarkeit an vorhandene Datenlieferungen (z.B. aus Finnova).

  • Die Verwendung von Parametertabellen erlaubt es, den InCube SA-CCR Calculator an zukünftige Veränderungen der Parameter (z.B. aufsichtsrechtliche Korrelationen, Volatilitäten, oder Skalierungsfaktoren) jederzeit flexibel und ohne grossen Aufwand anzupassen.

  • Die Berechnungsergebnisse (RWA und Kreditäquivalente) sowie die verwendeten Input-Daten können jederzeit zur Archivierung oder Weiterverwendung in nachgelagerten Systemen als .pdf oder .csv Dateien exportiert werden.

  • Der gesamte SA-CCR Calculator ist vollständig in einem Excel Workbook enthalten. Es ist keine Installation notwendig und es werden keine zusätzlichen Komponenten benötigt. Für weiterführende Informationen oder eine persönliche Vorstellung wenden Sie sich bitte an info@incubegroup.com

April 2017

FINMA bewilligt SIX und REGIS-TR als Transaktionsregister für das FinfraG-Reporting

Mit der FINMA-Aufsichtsmitteilung per 3. April 2017 informierte die FINMA über die Bewilligung eines schweizerischen Transaktionsregisters (SIX Trade Repository AG) sowie die Anerkennung eines ausländischen Transaktionsregisters (REGIS-TR S.A.).

Somit müssen gemäss FinfraG (Art. 104) / FinfraV (Art. 130) offene Derivatgeschäfte spätestens ab folgenden Zeitpunkten gemeldet werden:

  • ab dem 1. Oktober 2017, wenn die meldepflichtige Person eine Finanzielle Gegenpartei (FC), die nicht klein ist, oder eine zentrale Gegenpartei (CCP) ist;
  • ab dem 1. Januar 2018, wenn die meldepflichtige Person eine kleine Finanzielle Gegenpartei (FC-) oder eine Nichtfinanzielle Gegenpartei (NFC), die nicht klein ist, ist;
  • ab dem 1. April 2018 in den übrigen Fällen, wobei ein Geschäft zwischen zwei kleinen Nichtfinanziellen Gegenparteien (NFC-) nicht gemeldet werden muss.

Für Derivatgeschäfte, die über einen Handelsplatz oder ein Organisiertes Handelssystem (OHS) gehandelt werden, verlängern sich die Fristen um jeweils 6 Monate.

March 2017

Going Deeper in Finance

QuantAlea, a subsidiary of InCube Group, will be giving a talk at the NVIDIA Global Technology Conference 2017.

Summary of the talk

How wide is Deep Learning applicable in Finance? The talk starts with an overview of promising Deep Learning applications in Finance. We then focus on deep (variational), autoencoders show how they can learn hidden representations of unlabeled data and generate new data. This opens interesting new applications in anomaly detection, risk analysis, price prediction and algorithmic trading. We explore some of these use cases with real FX data and illustrate the concepts with interactive notebooks, showing how to build the models using frameworks such as Tensorflow and Keras, and how to use latest P100 GPUs for training.

Further information about GTC 2017 can be found at www.gputechconf.com.

March 2017

Prices Drop as You Shop – How Walmart is using Jet’s GPU based Smart Merchant Selection to Gain a Competitive Advantage

QuantAlea, a subsidiary of InCube Group, will be giving a talk at the NVIDIA Global Technology Conference 2017.

Summary of the talk

Last year Walmart acquired the New Jersey based startup Jet to improve their e-commerce platform with new innovations, to compete more successfully in the e-commerce market and to optimize their order fulfillment costs. A core value of the Jet platform is the smart merchant selection. When a customer orders several items at once they usually can be fulfilled from multiple merchants and different warehouses. The goal is to find the merchant and warehouse combination so that the total order cost, including shipment costs and commissions, is as low as possible. Jet developed an innovative solution to find the most attractive combination of merchants. Interestingly, the bigger the shopping cart, the larger the savings that can be generated. This talk explains how only a clever combination of Machine Learning, new algorithms and GPUs at scale in the cloud can address the problem. This allows to unlock new use cases and business applications, which would not be possible with traditional computing resources.

Further information about GTC 2017 can be found at www.gputechconf.com

March 2017

Basel III: Bestimmung der Kreditäquivalente von Derivaten («SA-CCR»)

Der Bundesrat und die FINMA haben die Eigenmittelverordnung (ERV) sowie das FINMA-Rundschreiben «Kreditrisiken – Banken» an die weiterentwickelten internationalen Normen nach dem Basel-III-Regelwerk angepasst. Wir geben eine Übersicht über die Kernpunkte der Bestimmung von Kreditäquivalenten für Derivate nach dem neuen Standardansatz («SA-CCR»), sowie die wichtigsten Änderungen im Vergleich zur Anhörungsversion sowie zur bisherigen Marktwertmethode.

Zudem stellen wir den InCube SA-CCR Calculator vor. Ein Excel-Tool zur einfachen und transparenten Berechnung der Kreditäquivalente von Derivaten gemäss SA-CCR sowie vereinfachtem SA-CCR vor.

November 2016

InCube implements stock screening app for Banque Cramer

Banque Cramer is getting in on the fintech wave: the Geneva-based private bank is launching Prime Selector, a stock screening app that was implemented by InCube and leverages the Bank's investment expertise. Prime Selector allows users to set a series of criteria on companies as well as to draw on Cramer’s investment research and ratings to set up portfolios. The tool can backtest what users’ strategies would have done in the last ten years, tracking performance, volatility, growth and best and worst monthly returns. 

The app's criteria are targeted towards a more sophisticated stock-picker than a typical retail client, with variants on key factors like growth, value and quality that are aimed towards financially savvy investors. Investors can for example activate a hedge overlay function when backtesting their strategy.

October 2016

Anpassungen der Anforderungen zur Corporate Governance bei Banken

Die FINMA informierte über ihr Vorhaben, die aufsichtsrechtlichen Anforderungen an die Corporate Governance, das interne Kontrollsystem und das Risikomanagement für Banken zu „straffen“ und zu „modernisieren“. Konkret geht es dabei um eine Totalrevision des Rundschreibens 2008/24 „Überwachung und interne Kontrollen Banken“ sowie um Teilrevisionen der Rundschreiben 2008/21 „Operationelle Risiken Banken“ und 2010/1 „Vergütungssysteme“. In der vorliegenden Publikation sind die wesentlichen Änderungen zusammengefasst.

August 2016

„Kreditrisiken Banken“ – Änderungen Eigenmittelverordnung und FINMA-Rundschreiben (Basel III)

Bundesrat und FINMA passen die Kreditrisikoregeln an die weiter entwickelten internationalen Basel-III-Normen an. Dazu haben sie Entwürfe der neuen Eigenmittelverordnung (ERV) und des FINMA-Rundschreibens „Kreditrisiken Banken“ veröffentlicht. Die vorliegende Publikation gibt eine Übersicht über die wesentlichen Änderungen. Ebenfalls stellen wir darin die Unterschiede zwischen der bisherigen Marktwertmethode (Current Exposure Method) und dem neuen Standardansatz für die Berechnung der Kreditäquivalente von Derivaten (SA-CCR) dar. Schliesslich stellen wir kurz unseren SA-CCR-Prototypen vor. Dieses Excel-Tool erlaubt uns, eine detaillierte Impact-Analyse durchzuführen, und es bietet wertvolle Unterstützung bei einer Implementierung und Umsetzung des neuen Standardansatzes.

March 2016

GPU Computing on .NET at Speed of CUDA C++

A Performance Comparison of Alea GPU C# and CUDA C++

Check out our new Blog

March 2016

Der neue Standardansatz zur Bestimmung des Kontrahentenausfallrisikos

In der vorliegenden Publikation untersuchen wir den neuen Standardansatz zur Bestimmung des Kontrahentenausfallrisikos und legen dabei besonderes Augenmerk auf die im Vergleich zur bisher häufig verwendeten Marktwertmethode deutlich angestiegenen Datenanforderungen. 

March 2016

“Auto-magic” GPU Programming

This blog is a preview of the new upcoming version 3 of Alea GPU which sets completely new standards for GPU development on managed platforms such as .NET or the JVM.

Check out our new Blog

March 2016

GPUs and Domain Specific Languages for Life Insurance Modeling

Check out Quantalea's new Insurance Blog:

Nvidia PARALLEL FORALL blog

QuantAlea Insurance Blog

March 2016

Advanced computation for intelligent investment

Marc Wildi and Boris Rankov outline their development of a forecasting system that supports decision making for asset allocation and portfolio management under variable financial market condition.

February 2016

Dokumentationspflicht FinfraG

Das Parlament hat im Sommer 2015 das "Bundesgesetz über die Finanzmarktinfrastrukturen und das Marktverhalten im Effekten- und Derivatehandel" (FinfraG) verabschiedet. Unser Merkblatt "Dokumentationspflicht FinfraG" fasst die wichtigsten Erkenntnisse zusammen.

December 2015

Herausforderungen an die Solvency II Reporting Datenqualität – Erkentnisse und Lösungsansätze aus der Vorbereitungsphase

Im vorliegenden Beitrag geht es darum, erste Erkenntnisse aus dem in der Vorbereitungsphase übermittelten quantitativen Berichtswesen an die zuständigen Regulatoren vor dem Hintergrund der qualitativen Anforderungen an die Daten zu analysieren und Lösungsvorschläge zur Verbesserung zu präsentieren.

December 2015

Die neue Standardmethode zur Behandlung des Gegenparteiausfallrisikos (SA-CCR)

Der Basler Ausschuss für Bankenaufsicht hat eine neue Standardmethode zur Bestimmung des Gegenparteiausfallrisikos von ausserbörslich und börsengehandelten Derivaten sowie Geschäften mit langer Abwicklungsfrist erarbeitet. Diese ersetzt per Anfang Januar 2017 die bisherige Standardmethode sowie die Marktbewertungsmethode.

October 2015

F# for Industrial Applications – Worth a Try?

F# is a functional first language. It is cross platform, open-source with professional development tools and backed by an active community. The popularity of F# steadily increased. It is used in specialized areas such as scientific computing, data science, machine learning, or for trading and risk applications.

September 2015

FinfraG Update: Release of FinfraV

On 20th of August 2015, the ordinance of FinfraG (FinfraV) was released, and the consultation period will last until 2nd of October 2015. For the first time, the implementation details are known in a draft status (counterparty classification thresholds, risk reduction obligations).

September 2015

Dr. Daniel Egloff at the Jazoon Conference Oct. 23rd 2015

Dr. Daniel Egloff will hold two speeches at the Jazoon TechDays, October 23rd 2015, at Technopark in Zurich Switzerland.

The goal of the conference is to give developers a profound insight into different topics about programming languages and emerging “fashionable” technologies besides Java.

At 10.20 a.m.: .NET Applications with GPUs and the cloud

At 1.20 p.m.: F# for industrial applications - worth a try?

Please find the detailed agenda and register for your ticket directly here:

Jazoon Website

September 2015

Solvency II: InCube Assets QRT Validator

During the production of the June 2015 Solvency II QRT interim measures submission, lacking data quality (completeness, validity and consistency) turned out to be an issue for several insurance companies. To establish a robust production process data quality needs to be validated as early as possible such that potential issues can be immediately resolved.

The InCube Assets QRT Validator tests whether the investment data provided for the QRT fulfils the EIOPA requirements before submitting to any potential reporting solution and the final XBRL generation.

August 2015

IFRS 9 "Financial Instruments" - part 2 "impairments"

In the first issue of this publication series on IFRS 9 we have covered the classification and measurement issues as well as the interactions with IFRS 4 Phase 2 (see February 2015).

In this paper we focus on impairments. The new impairment model of IFRS 9 is based on expected credit losses rather than incurred losses. It is therefore expected to result in the earlier recognition of credit losses. Insurers with large volumes of amortized cost and FVTOCI debt instruments will see the biggest impact, particularly on transition from IAS 39 to IFRS 9.

June 2015

Switzerland adopts international standards for financial market infrastructures

The Council of States and the National Council have intensely discussed their remaining differences regarding FinfraG this week. At the end the parliament came to an agreement on all remaining open issues.

June 2015

Entry into force of the partially revised Ordinance on the Supervision of Private Insurance Companies

On 25 March 2015, the Federal Council decided to bring the partially revised Insurance Supervision Ordinance (ISO) into force on 1 July 2015. This will result in changes in the supervision of private insurance companies and of insurance groups and conglomerates.

Solvency, qualitative risk management and disclosure in particular will undergo substantial changes. Adjustments will also be made to technical provisions, tied assets and the supervision of intermediaries, and some branch-specific provisions.

March 2015

Alea GPU released

Alea GPU, the first professional GPU software development environment for .NET, has been successfully released. 

Alea GPU, a truly cross-platform product, allows you to write GPU code once and run it on Windows, Linux or OS X. It improves productivity with first class tooling for coding, debugging and profiling and reduces development time with prefabricated GPU algorithms and specialized libraries. Alea GPU further increases agility with GPU scripting and a rapid prototyping REPL.

Get more information on Alea GPU by visiting the QuantAlea website, or directly download the Alea GPU packages on nuget:

QuantAlea website

Download Alea GPU on nuget

QuantAlea is a company of InCube Group.

March 2015

QuantAlea has launched its new website

QuantAlea, a company of InCube Group, has launched its new website. The company's flagship product, AleaGPU, provides the first professional GPU software development environment for .NET.

Visit the new website here:

QuantAlea

February 2015

IFRS 9 „Financial Instruments“ – important aspects for insurance companies

In this first issue on IFRS 9 we look at the classification and measurement of financial instruments under this new accounting standard. We provide an overview of the requirements and highlight the important aspects for insurance firms.

IFRS 9 requires from insurers decisions on the classification and measurement of financial instruments that may have large impacts on their financial results. Particularly, impacts that may result from the interaction between IFRS 9 and IFRS 4 Phase 2 “Insurance Contracts” have to be closely analyzed.

January 2015

Fundamental Review of the Trading Book; Part 2

This is the second issue in our series on the Basel initiative to fundamentally revise the trading book regulation known as “Fundamental Review of the Trading Book”. We present the new internal models-based approach, highlight its most important changes and features, and discuss major industry challenges in the light of the latest developments in the new set of rules.

(To read Part 1 of “Fundamental Review of the Trading Book”, please click here)

November 2014

Solvency II: Preparatory Phase - What needs to be done next?

Early preparation is essential for meeting the new requirements that will apply when Solvency II becomes effective. The presentation summarizes the current status and provides an outlook on the next steps which need to be performed.

November 2014

Code Mesh London 2014: Taming GPU threads with F#

Dr. Daniel Egloff held a talk at the Code Mesh London on 5 November 2014.

Abstract:

Writing GPU kernel code which optimally exploits parallelism and the GPU architecture is the most challenging and time-consuming aspect of GPU software development. Programmers have to identify algorithms suitable for parallelization and while implementing them reason about deadlocks, synchronization, race conditions, shared memory layout, plurality of state, granularity, throughput, latency and memory bottlenecks.

New languages and tools to increase the productivity in GPU software development, whilst at the same time retaining the full flexibility of the underlying programming models such as CUDA or OpenCL, are thus of tremendous value.

In this talk, Daniel gave an introduction to InCube's high productivity GPU development tool chain Alea GPU based on F#. He showed how GPU scripting, dynamic compilation and unique features of the F# language can be leveraged to reduce the development time of reusable libraries of parallel primitives and core numerical algorithms.

October 2014

Daniel Egloff visiting Microsoft Research, Cambridge UK

We are delighted that Microsoft Research in Cambridge, UK has invited Dr. Daniel Egloff, Partner at InCube Group, to meet with Don Syme and give a lecture on ‘Taming GPU threads with F# and Alea.GPU’. The lecture is scheduled at Microsoft Research Cambridge on Nov 3, 2014.

Don Syme is a Principal Researcher at Microsoft Research, Cambridge, UK. He is the designer and architect of the F# programming language.

Microsoft Research employs computer scientists, physicists, engineers, and mathematicians to collaborate with academic, government, and industry researchers to advance the state of the art of computing, and solve difficult world problems through technological innovation. 

For more information on Microsoft Research visit: 

Microsoft Research  

October 2014

Operational risk management - Risk prevention when dealing with electronic client identifying data

To strengthen the confidentiality of client identifying data (CID) the Swiss Financial Market Supervisory Authority FINMA has published Appendix 3 of the FINMA Circular 2008/21 (Operational risks at banks) which will become effective on January 1st, 2015. In this publication we are describing the major challenges arising during the implementation of Appendix 3.

October 2014

CTI Project: Multi-period portfolio construction with macro-economic turning-point signals

We are pleased to announce that the Commission for Technology and Innovation (CTI) has accepted a joint research proposal by InCube Group AG, Zurich University of Applied Sciences Winterthur (ZHAW) and University of Sydney. The research project will be jointly funded by InCube Group AG and the CTI and will last 29 months.

More information on the CTI can be found here:

Commission for Technology and Innovation (CTI)

September 2014

FinfraG / EMIR - What you need to know

Please find below the slides presented at the FinfraG / EMIR conference on 23 September 2014. 

September 2014

Taming GPU threads with F#

Dr. Daniel Egloff will hold a talk at the Code Mesh London on 5 November 2014.

Abstract:

Writing GPU kernel code which optimally exploits parallelism and the GPU architecture is the most challenging and time-consuming aspect of GPU software development. Programmers have to identify algorithms suitable for parallelization and while implementing them reason about deadlocks, synchronization, race conditions, shared memory layout, plurality of state, granularity, throughput, latency and memory bottlenecks.

New languages and tools to increase the productivity in GPU software development, whilst at the same time retaining the full flexibility of the underlying programming models such as CUDA or OpenCL, are thus of tremendous value.

In this talk we give an introduction to our high productivity GPU development tool chain Alea based on F#. We show how GPU scripting, dynamic compilation and unique features of the F# language can be leveraged to reduce the development time of reusable libraries of parallel primitives and core numerical algorithms.

For more information visit: 

Code Mesh London 2014

September 2014

Fundamental Review of the Trading Book; Part 1

This is the first issue in our series on the Basel initiative to fundamentally revise the trading book regulation known as “Fundamental Review of the Trading Book”. We provide an overview of its history, content, and objectives and highlight major industry challenges and latest developments.

September 2014

InCube Group and QuantAlea strengthen their market position

After years of close, highly successful cooperation InCube Advisory, InCube Capital and QuantAlea have decided to merge their businesses into a single unit.

August 2014

What you need to know about FinfraG/EMIR

Seminar on OTC derivatives regulations according to the Swiss Financial Market Infrastructure Act (FMIA/FinfraG) and the European Market Infrastructure Regulation (EMIR):

Tuesday, 23 September 2014, 3.30 p.m.

Hotel Schweizerhof, Bahnhofplatz 7, Zurich

Guest speaker: Reto Ziltener, Head Prime Services, Zuercher Kantonalbank

March 2014

How to Design a Language Integrated Compiler with LLVM

Xiang Zhang, the lead developer of QuantAlea, and Aaron Brewbaker explained in their talk at NVIDIA's GTC conference in San José, California, how to design a language integrated compiler with LLVM and illustrated the concepts with concrete implementations from Alea.cuBase.

March 2014

GPU Applications for Modern Large Scale Asset Management

Dr. Daniel Egloff gave a talk at NVIDIA's GTC conference in San José, California, about GPU applications for modern large scale asset management.

February 2014

Qualitative liquidity risk requirements – proposed changes

Overview of proposed changes to the Liquidity Ordinance and FINMA Circular 2013/06 “Liquidity Banks” in Switzerland (implementation of Basel III liquidity risk).

February 2014

Basel 3.5: Fundamental Review of the Trading Book

Summary of the 2nd consultative document on the fundamental review of the trading book (revised market risk framework) of the Basel Committee on Banking Supervision issued in October 2013.

January 2014

OTC derivatives regulation – Financial Market Infrastructure Act (FMIA): interview with Handelszeitung

Article “Vertrackte Kontrakte” published by the Handelszeitung (Switzerland) based on an interview with Erich Felder, CFA.

October 2013

F# and GPUs: speed, speed, speed!

Dr. Daniel Egloff will be talking at the upcoming F# in Finance all day events in London and New York.

Abstract

In the fast moving world of financial markets it's speed, speed, speed! GPU compute accelerators are finding wide application in pricing and risk, and wherever else speed is a compatitive advantage. Now its easy to tap into GPU compute power with F# and Alea.cuBase.

For more information visit: 

New York F# User Group

F# in Finance New York, December 2013

September 2013

Large Scale Optimization and High Performance Computing for Asset Management

Dr. Daniel Egloff will present at the GPU Club of the University of Manchester on Wednesday, 2 Oct 2013.

For more information visit:

University of Manchester Computational Science Community

June 2013

finews.ch: interview with Erich Felder, CFA

Interview organized by finews.ch, a Swiss online news portal, on regulatory developments and their impact on independent asset managers in Switzerland. Topics: US tax deal, FATCA, AIFMD, Basel III, EMIR, Dodd-Frank, FinfraG, MiFID, MiFIR, FIDLEG, correlation breakdown, risk-factor approaches.

June 2013

Presentation on Gold investments at the annual Commodity Club Gold Event

Lorenz Arnet and Boris Rankov from InCube presented at the annual Gold Event organized by the Commodity Club Switzerland (www.commodityclub.ch) on the topic "Active Gold Strategies designed to outperform the Gold price".

June 2013

OTC derivatives reform: European Market Infrastructure Regulation (EMIR)

The EU has introduced the European Market Infrastructure Regulation (EMIR) following the G20 commitments to improve the efficiency and transparency of the OTC derivatives markets with a view to reducing counterparty, operational risk and overall systemic risk. This document gives an introduction to EMIR and to the situation for counterparties in Switzerland.

May 2013

.NET in the City - F# and Computational Finance

The last .NET in the City event organized by Microsoft last Thursday, May 16th, was a great success. We had more than 70 registrations and a large crowed showed up.

Daniel Egloff gave another presentation about F# and its application to computational finance and GPU computing. He showed several live coding examples, including GPU scripting inside Excel, as well as interoperation of CUDA with graphics visualization through DirectX.

Don Syme was present as well and tweeted directly from Daniel's talk. Shortly after Sergey Tihon listed Don's tweets on this blog post.

April 2013

F# + GPGPU - Financial Services Developer Forum (.NET in the City)

Join the Daniel Egloff at the .NET in the City event taking place May 16th.

This exclusive Microsoft event is aimed at Developers and Architects from the Financial Services sector, providing an opportunity to learn about some of the new developer capabilities, and best practices.

For more information visit:

F# + GPGPU - Financial Services Developer Forum

April 2013

From Parallel Algorithms to Monads - New Techniques for Using GPUs to make Derivative Pricing and Risk Analytics more effective

Dr. Daniel Egloff gave a presentation at the 2013 ICBI Global Derivatives Trading & Risk Management conference in Amsterdam about how computational workflows or monads can be used to simplify reusability and composition of GPU algorithmic building blocks to more complex GPU algorithms, thereby significantly reducing time to market, development cost and at the same time improve the correctness and robustness of the application.

March 2013

Dynamic CUDA with F# – New Dimensions for GPU Computing on .NET

Dr. Daniel Egloff and Xiang Zhang presented Alea.cuBase at GTC 2013 in San José and showed how to do GPU scripting with F# and Excel.

For some other opinion see also the comments on the AccelerEyes blog.

March 2013

Alea.cuBase and F# - a new level of GPU Computing

Dr. Daniel Egloff will speak at the meet up of the HPC & GPU Supercomputing Group of Silicon Valley, taking place Tuesday March 19 at the San Jose Convention Center, Room 210F.

For more details go to HPC & GPU Silicon Valley group or to The San Francisco Bay Area F# user group.

March 2013

GPU Computing in Finance with Alea.cuBase

Daniel Egloff gave a brief introduction to GPU computing for financial applications with Alea.cuBase.

Below you can download the slides and the demo project. In order to run the demo code you have to install either a trial version of Alea.cuBase or request a personal license directly from QuantAlea with a mail to info(at)incubegroup.com.

March 2013

Basel III: new liquidity risk requirements for banks

Overview of the key points of the new Basel III liquidity risk requirements introduced with the Liquidity Ordinance and FINMA Circular 2013/06 “Liquidity Banks” as well as their implications and challenges for banks.

February 2013

GARP Zurich Chapter presentation: "Consistent risk management with multipurpose pricing infrastructures"

Slides presented by Dr. Daniel Egloff at the GARP Zurich Chapter meeting on 5 February 2013. They cover approaches on how to overcome the challenges in the front, middle and back offices of financial institutions: (1) requirement for consistent representation, valuation and management of financial products for all purposes; (2) need to adopt new technologies, such as GPUs, clusters, grids, clouds, virtual machines; and (3) requirement to integrate the two given the large amount of legacy code.

February 2013

Introducing Alea.cuBase

In order to respect existing trademarks, we decided to change the product name from Alea.CUDA to Alea.cuBase and removed all previous releases of Alea.CUDA from our web site. Existing Alea.CUDA licenses will not work anymore for Alea.cuBase.

Licensed users will find new license keys in their user account on our web site. The name change comes with a new release. Besides some performance improvements in the launch functions we removed the examples and the tutorials from the release. They are now available on GitHub as independent publicly available projects, which allows us to upgrade the examples independently of a release.

For more details please check out the release notes of Alea.cuBase.

January 2013

Free Alea.cuBase Licenses for Participants of F# User Group Meetings Related to Alea.cuBase

QuantAlea offers free license to the F# user group community for a limited period of time.

To request a free personal license please fill out the license request form. To qualify for a license you have to be a registered user of an F# user group and you must have attended a user group meeting related to Alea.cuBase or our GTC 2013 talk.

QuantAlea reserves the right to reject a license request if the supplied information is incomplete or insufficient. We also reserve the right to withdraw this offering at any time. 

January 2013

GARP Chapter Meeting, Zürich, Switzerland

Consistent Risk Management with Multipurpose Pricing Infrastructures

Synopsis:

Dr. Daniel Egloff will discuss designing and implementing pricing infrastructures, which can be used for multiple purposes including front office pricing, risk management, structuring, or ad-hoc analytics, such as hedging efficiency analysis.

A multipurpose pricing infrastructure is far more than a quant library deployed on a few multi-core boxes. Rather it is about suitable abstractions and unifying concepts, proper model implementation, efficient use of modern hardware technologies such as Graphics Processing Units (GPU), the selection of the right engineering tools, and deployment to high throughput cluster and cloud infrastructures.

These points will be illustrated by means of a concrete example for equity basket derivatives and the local volatility model. Various cases will be explored to demonstrate the benefit of a holistic approach to pricing and risk management, from a quant modelling, software engineering and hardware infrastructure point of view.

Registration:

You can register for the presentation followed by reception free of charge directly on the GARP web site.

January 2013

Dynamic CUDA with F# - New Dimensions for GPU Computing on .NET

We will be presenting at the NVIDIA GTC conference, taking place March 18-21 at the San Jose McEnery Convention Center in San Jose, California.

Abstract:

CUDA and F# are two trailblazing yet unrelated technologies. F# is a uniquely productive language to solve complex problems in a clear and concise way. On the other hand CUDA matured to an industry standard for parallel high performance computing on GPUs.

Our presentation shows how Alea.cuBase combines the two technologies F# and GPUs, leading to a complete solution to develop GPU accelerated applications on .NET. Alea.cuBase builds upon the LLVM compiler toolkit and the new NVIDIA PTX backend of CUDA 5. With Alea.cuBase you effectively write F# code, which generates CUDA programs dynamically at runtime, fully integrated into .NET.

January 2013

Successful start of the InCube Gold Fund

We are pleased to announce the successful start of the InCube Gold Fund, outperforming its benchmark by 1.28% in the first month of trading.

December 2012

Basel 2.5 and Basel III: a brief guide through the maze

The financial crisis has highlighted the need to reform the Basel II capital framework and to complement it with a liquidity risk framework. This document summarizes the main revisions and enhancements introduced with Basel 2.5 and Basel III, the key documents issued by the Basel Committee on Banking Supervision and the Swiss regulator, and the implementation dates.

November 2012

Introducing Alea.CUDA at Zurich F# Users

The meetup will be about F# computation expressions (also referred to as workflow expressions or monads) in practice. I will begin with a brief overview. After that, James Litsios will present an example related to modular pure state management. Finally, Daniel Egloff will present a case study related to Alea.Cuda, a new F# product for processing on the GPU, which Daniel's company, QuantAlea, has created in the context of a client project.

For more information visit:

Zurich F# Users

November 2012

Introducing Alea.CUDA at Functional Londoners

F# and GPUs are two trailblazing yet unrelated technologies. F# is a uniquely productive language to solve complex problems in a clear and concise way. On the other hand GPUs offer an immense computational power to solve large number crunching tasks fast and efficiently.

Our presentation shows how to wed the two technologies F# and GPUs with the help of Alea.CUDA. Alea.CUDA is our new framework and compiler service for GPU computing. It extends F# with the key CUDA concepts and allows to compile F# code quotations to an executable GPU code. I will briefly introduce Alea.CUDA and show you – by means of several live coding examples – how it can be used to develop GPU algorithms entirely in F# with the full flexibility of CUDA-C.

Besides getting an understanding of the main features of Alea.CUDA you will become familiar with some of the basic GPU computing paradigms. To round off the presentation I shall reveal some of the implementational aspects of Alea.CUDA.

Podcast

July 2012

EMIR: new challenges for financial institutions

Overview of the requirements introduced with the European Market Infrastructure Regulation (EMIR) and their challenges to financial institutions.

May 2012

New Generation GPU Accelerated Financial Quant Libraries

Presenting "New Generation GPU Accelerated Financial Quant Libraries" at NVIDIA Global Technology Conference 2012, San José, California.

New generation GPU accelerated solutions for derivative pricing, hedging, and risk management can be build more efficiently with modern technology and functional programming languages like F# on .NET or Scala on the Java VM. As a concrete example we report from a large derivative pricing project developed in F# on .NET.

We will introduce the key design concepts and parallelization strategies, which lead to an efficient and transparent GPU acceleration. Several examples will illustrate the benefit of functional as compared to the classical object oriented approach.

March 2012

Pricing Financial Derivatives with High Performance Finite Difference Solvers on GPUs

Abstract:

The calculation of the fair value and the sensitivity parameters of a financial derivative requires special numerical methods, which are often computationally very demanding. In this chapter we discuss the design and implementation of efficient GPU solvers for the partial differential equations (PDEs) of derivative pricing problems.

For derivatives on a single asset like a stock or an index we consider a massively parallel PDE solver which simultaneously prices a large collection of similar or related derivatives with finite difference schemes. We achieve a speedup of a factor of 25 on a single GPU and up to a factor of 40 on a dual GPU configuration against an optimized CPU version.

Often derivatives are written on multiple underlying assets, e.g. baskets, or the future asset price evolution is modeled with additional risk factors, like for instance stochastic volatilities. The resulting PDE is defined on a multidimensional state space. For these kind of derivatives it is not necessary to pool multiple pricing calculations: alternating direction implicit (ADI) schemes for PDEs on two or more state variables have enough parallelism for an efficient GPU implementation. We benchmark a specific ADI solver for the Heston stochastic volatility model against a fully multi-threaded and optimized CPU implementation. On a recent C2050 Fermi GPU we attain a speedup of a factor of 70 and more for a sufficiently large problem size.

Our results demonstrate the importance of the effective use of GPU resources such as fast on-chip memory and registers.

For more information visit: 

GPU Computing Gems Jade Edition (Applications of GPU Computing Series), Wen-mei W. Hwu (Editor), 2012, p. 309-322

November 2011

Feeling partial: tackling the challenges of PDEs in the brave new world of GPUs in finance

Wilmott Magazine interview with Dr. Daniel Egloff. Beyond huge advantages in terms of efficiency and long-term costs the steady adoption of GPU architecture in finance is bringing with it a shift, not only in the way the industry thinks about hardware, but also the thought given to the logical flow of the computational process and, potentially, the kinds of models utilized in quantitative finance.

March 2011

GPUs in financial computing part III:

The paper looks at the design and implementation of efficient GPU solvers for two factor models with a focus on stochastic volatility models. The resulting partial differential equations of two state variables are solved with alternating direction implicit (ADI) schemes. The paper shows that ADI-style schemes can be parallelized very efficiently on a GPU. Already a single pricing problem can utilize the full GPU capacity, a pooling of multiple pricing calculations to generate more parallelism is not necessary anymore.

Wilmott Magazine, March 2011

December 2010

American options with stopping time constraints

This paper concerns the pricing of American options with stochastic stopping time constraints expressed in terms of the states of a Markov process. A transformation is applied to express the contract as a generalized barrier option. The valuation problem of the barrier option leads to a stochastic Cauchy-Dirichlet problem which we numerically solve with a suitable extension of the Longstaff-Schwartz algorithm.

Applied Mathematical Finance, Volume 16, Issue 3, 2009, p. 287-305.

November 2010

GPUs in financial computing part II: massively parallel PDE solvers on GPUs (Wilmott Technical Paper)

Computing a large number of option prices with finite difference schemes in parallel leads to enough data-parallel work to fully load a modern GPU. The paper proposes a massively parallel PDE solver which simultaneously prices a large collection of similar or related derivatives with finite difference schemes. The performance analysis shows a remarkable speedup by a factor of 25 on a single GPU and up to a factor of 40 on a dual - GPU configuration against an optimized CPU version.

Wilmott Magazine, November 2010

October 2010

The term structure of variance swap rates and optimal investing in variance swaps

We study the optimal investment decision on the variance swap contract and the stock index. We find that two factors are needed to capture the term structure variation of the variance swap rates. Also the presence of variance swap contracts alters the investor’s optimal portfolio decision.

Journal of Financial and Quantitative Analysis, volume 45, issue 05, 2010, pp. 1279-1310.

September 2010

GPUs in financial computing part I: high-performance tridiagonal solvers on GPUs (Wilmott Technical Paper)

The financial industry starts to adopt GPUs more and more. The main application fields as of today are Monte Carlo simulations. They fall into the category of embarrassingly parallel problems and can be implemented on a GPU in a relatively straightforward manner, once a good random number generator is available. Speedups of a factor of 50 to 100 are within reach.

Wilmott Magazine, September 2010

March 2010

Mind your language - Python takes a bite

Interview with Dr. Daniel Egloff in the Wilmott Magazine on the use of Python in financial applications and software architectures.

August 2009

Quantile estimation with adaptive importance sampling

We introduce new quantile estimators with adaptive importance sampling. The adaptive estimators are based on weighted samples that are neither independent nor identically distributed. Using a new law of iterated logarithm for martingales, we prove the convergence of the adaptive quantile estimators for general distributions with nonunique quantiles, thereby extending the work of Feldman and Tucker. We illustrate the algorithm with an example from credit portfolio risk analysis

April 2009

American options with stochastic stopping time constraints

This paper concerns the pricing of American options with stochastic stopping time constraints expressed in terms of the states of a Markov process. We apply a transformation to express the contract as a generalized barrier option. The valuation problem of the barrier option leads to a stochastic Cauchy-Dirichlet problem which we numerically solve with a suitable extension of the Longstaff-Schwartz algorithm.

Applied Mathematical Finance, Volume 16, Issue 3, 2009, p. 287-305

April 2008

Teraflops for Games and Derivative Pricing

Financial computing continuously demands higher computing performance, which can no longer be accomplished by simply increasing clock speed. Cluster and grid infrastructures grow, their cost of ownership explodes. On the other hand, the latest GPU (Graphics Processing Unit) boards show impressive performance metrics. This leads to the questions if and how one can harness this power to bring financial computing to the next level. We analyze the pricing of equity basket options with a local volatility model implemented on a GPU. Our performance gains are very impressive.

January 2007

A dynamic look-ahead Monte Carlo algorithm for pricing Bermudan options

Under the assumption of no-arbitrage, the pricing of American and Bermudan options can be casted into optimal stopping problems. This paper proposes a new adaptive simulation based algorithm for the numerical solution of optimal stopping problems in discrete time.

Annals of Applied Probability Volume 17, Number 4, 2007, p. 1138-1171.

February 2006

A simple model of credit contagion

The paper proposes a simple model of credit contagion in which macro- and micro-structural interdependencies among the debtors within a credit portfolio are included. The microstructure captures interdependencies between debtors that go beyond their exposure to common factors, e.g., business or legal interdependencies. We show that even for diversified portfolios, moderate micro-structural interdependencies have a significant impact on the tails of the loss distribution. This impact increases dramatically for less diversified microstructures.

Journal of Banking & Finance, Volume 31, Issue 8, August 2007, Pages 2475–2492.

This paper won the STOXX 2004 Gold Award of the annual meeting of the European Financial Management Association.

May 2005

Monte Carlo algorithms for optimal stopping and statistical learning

We extend the Longstaff–Schwartz algorithm for approximately solving optimal stopping problems on high-dimensional state spaces. We reformulate the optimal stopping problem for Markov processes in discrete time as a generalized statistical learning problem. Within this setup we apply deviation inequalities for suprema of empirical processes to derive consistency criteria, and to estimate the convergence rate and sample complexity. Our results strengthen and extend earlier results obtained by Clément, Lamberton and Protter [Finance and Stochastics 6 (2002) 449–471].

Annals of Applied Probability Volume 15, Number 2, 2005, p. 1396-1432.

2015

2014

2013

2012 and older