S&P Capital IQ Portfolio Risk Webinar Series
Re-Thinking
Scenarios
Analytics and Stress Testing for Multi-Asset Portfolios Next Generation of Scenario
Dr. Dan Rosen Managing Director, Risk and Analytics S&P Capital IQ March 5th, 2014 Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public. Copyright © 2014 Standard & Poor’s Financial Services LLC, a subsidiary of The McGraw –Hill Companies, Inc. All rights reserved.
CHARTERED ALTERNATIVE INVESTMENT ANALYST (CAIA) ASSOCIATION The Global Benchmark In Alternative Investment Education
SHOWCASE YOUR KNOWLEDGE
ABOUT CAIA Our Mission Establish the CAIA designation as the benchmark for alternative investment education worldwide Promote professional development through continuing education, innovative research and thought leadership Advocate high standards of professional ethics Connect industry professionals globally Our Program: A self-study program. Two exams offered twice a year Our Members:
6,700 Charter Holders 17 Chapters Around the World
GLOBAL PRESENCE
September 2014 Exam.
MEMBERSHIP PROFILE
EMEA 32%
APAC 13% September 2014 Exam.
AMRS 55%
CAIA CURRICULUM Level I
Level II
Professional Standards & Ethics
Professional Standards & Ethics
Introduction to Alternative Investments
Private Equity/VC
Real Assets
Real Assets
Hedge Funds
Commodities
Commodities
Structured Products
Private Equity
Due Diligence
Structured Products Risk & Portfolio Management
Hedge Funds/Managed Futures
Risk Management Asset Allocation & Portfolio Management
S&P Capital IQ Portfolio Risk Webinar Series
Re-Thinking
Scenarios
Analytics and Stress Testing for Multi-Asset Portfolios Next Generation of Scenario
Dr. Dan Rosen Managing Director, Risk and Analytics S&P Capital IQ March 5th, 2014 Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public. Copyright © 2014 Standard & Poor’s Financial Services LLC, a subsidiary of The McGraw –Hill Companies, Inc. All rights reserved.
Webinar Series: Re-Thinking Models And RISK Analytics
In this webinar series, we discuss the evolution, best practices and recent research that is now transforming the way financial models and analytics are applied to manage multi-asset portfolios.
8
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Webinar Series: Re-Thinking Models And RISK Analytics
In long periods of benign conditions: Fading memory can lead to complacency and underpricing of risk In periods of expansion: Innovation and new products, rapid growth, limited/no loss data
9
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
This Webinar – It’s All About Scenarios • Scenario analysis and stress testing have been an explicit part of risk management methodologies and systems for a while • But the typical scenario analysis tools are generally quite static, largely manual,
incomplete and dated • Technology, data and models are now available to take scenario analysis to the next level in practice… • This is happening in other businesses, as well as physical & social sciences 10
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios‌
Scientific American December 2011 Source: Scientific American, December 2011. For illustrative purposes only.
11
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios And Predicting The Future
Source: http://www.futurict.eu/ - For illustrative purposes only.
12
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Introduction: Scenario-Based Risk And Performance Paradigm Nirvana: integrated, actionable view of
performance and risk for multi-asset portfolios in real time
13
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Scenario-Based Risk And Performance Paradigm – Four Pillars
The design and testing of strategies, reactions and intervention measures (management, policy modelling and governance) 14
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Scenarios – What Next To What If
The quality of risk analysis and the actions/decisions will rely on our ability to generate relevant, plausible and comprehensive, forward-looking scenarios that properly represent the future 15
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Scenarios – What Next To What If
• Scenarios provide efficient extrapolations from real-world situation • Produce warnings of critical developments
• Determine causal interdependencies which facilitate to identify likely or possible next events • Discover effects of structural change in a system, complex interconnections, emergent and reflexive behavior, and non-linear responses, possibly resulting in self-organized criticality 16
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
What Is A Scenario? • A scenario is the basic descriptor of the evolution of the state-of-theworld over time – It gives a joint realization of all the relevant financial and economic risk factors at a point in time or, more generally, a discrete set of times in the future
• Scenario generation differs considerably from forecasting – A forecast is a prediction that a single scenario will occur, and its accuracy is therefore crucial – But… we really can’t predict accurately specific financial events in the future
17
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Scenarios And Modelling The Future
“Analysts successfully predicted 12 out of the last 4 major recessions…” (Samuelson 1991)
18
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Different Types Of Scenarios • Sensitivity scenarios – Key bucket shifts; PV01, CR0; parallel shifts; discrete factor scenarios; JtD; convexities, cross-gammas…
• Subjective/expert scenarios – Basic (examples: curve parallel shift, twist etc. slides of shifts on a factor or set of these, etc.) – Research and analyst reports – Expert out-of the-box scenarios (e.g., asking risk managers, or from the news, etc.)
• Historically-based scenarios – Crises, smart searches for specific periods or characteristics (conditional scenarios)
• Regulatory/compliance scenarios – Banking regulators, government stress tests, industry reporting standards
• Statistical scenarios – Historical market scenarios (various periods, searches, and weighting schemes)
e.g., stressed period, normal period (non-parametric methods) – Monte Carlo (market risk) – from joint Normal to general copula representations
SPY
116
114
112
– Credit Risk: credit migration and default scenarios – Robust MC and model risk…
• Implied views (reverse stressed testing, inverse problems)
110
108
10:00
Source: S&P Capital IQ - For illustrative purposes only
19
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
12:00
14:00
16:00
Re-Thinking Scenarios – Seven New Principles… 1. Scenarios are not about the risk factors…. • They are about – How these factors affect our portfolios – Actions we can take to make better investment decisions and/or manage risk
• It is generally hard to apply scenarios to multiple portfolios across the firm, and share them within the firm or externally More efforts across the industry will go into this…
20
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios 1. Scenarios are not about the risk factors… they are about portfolios
2. Scenarios tell stories – we need to become much better story-tellers – Spend more time understanding the meaning and the consequences – primary and secondary characters, sub-plots
– Much better interactive/live reporting – remember, its about the portfolio – Complementing scenario analytics with qualitative information, deep instrument and obligor dives (financials, business and management descriptions) analysts reports, context, news, … – Multiple views and sources…
21
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Historical Crisis Scenarios 45
S&P 500
VIX
1100
40
• Black Monday (1987)
35 1050
• Asian Currency Crisis (1997)
30
25
• LTCM (1998)
1000 DJI
20
• Dot-com bubble crash (2001)
Sep 03
Sep 17
Oct 01
2600.
Oct 15 Jan
2400.Apr
• 9-11 (2001)
Oct
Jan
2200. 1400
• Great recession crashes (2008)
2000.
S&P 500
1300
– Bear Sterns – Lehman
Jul
1800.
1200 SPY
116
5000
NASDAQ
Jul
1100
Sep
Nov
114
• Flash Crash (2010)
4000
1000 112
• Debt Ceiling (2011)
900
3000 110
800
2000
• Fiscal Cliff (2012)
Jul
Sep
108
1000 10:00
Source: S&P Capital IQ - For illustrative purposes only
22
0 1990
1995
2000
2005
12:00
14:00
16:00
2010
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Nov
Jan
Jan
Example: Historical Crisis Scenarios • Dot-com… the plot – During the 1990’s low interest rates, as well as a “technology” hype drew investors towards the internet sector. During this period several Internet based companies were founded with the expectation that they would become profitable at a later time. Unfortunately, the growth was unsustainable and in the year 2000 this bubble burst, one of the triggers for this burst is attributed to the released from the US federal court stating the Microsoft was a Monopoly. The bubble deflated between 2000 and 2001, by the time it finished deflating several of those Internet start-up companies had completely disappeared. 5000 NASDAQ
4000
3000
2000
1000
0 1990
1995
2000
2005
2010
6.5
6.0
5.5
5.0
February 1, 2001
3 0Y
7Y
1 0Y
4Y 5Y
3Y
2Y
1Y
6M
3M
April 3, 2001
USD Swap Rate Curve
Source: S&P Capital IQ - For illustrative purposes only.
23
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Historical Crisis Scenarios • But… remember (Principle 1):
The story is NOT about the economy or the risk factors… It’s all about the PORTFOLIO and how it is impacted on these scenarios
24
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios 1. Scenarios are not about the risk factors… they are about portfolios
2. Scenarios tell stories – we need to become much better story-tellers
3. Scenarios must be comprehensive and robust – They must describe all factors that affect our portfolios – Scenario levels must be accurate and relevant Economically For the business user’s portfolio
25
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Incomplete Scenarios And Filling The Gaps • Scenarios are naturally described in terms of a small number of key risk factors (the main characters of the plot) • When applied to a given portfolio: they are incomplete – they may not describe what happens to all relevant risk factors that affect it • Implicitly or explicitly this is solved when the scenario is applied to the portfolio – e.g., other risk factors are ignored or given simple subjective views
• Theoretical, statistical or expert-based models on the joint behaviour of the factors can be used to create “complete scenarios” – Single scenario or conditional distributions
26
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Incomplete Scenarios And Filling The Gaps • Models on the joint behaviour of the factors can be used to create complete “single scenarios” or “conditional distributions” • Methods include – Explicit proxying based on expert knowledge or grouping methods (e.g., ratings and sector…) – Rules-based systems applied to a given portfolios (e.g., shock by 50bps all HY CDSs in the construction sector, but 60bps those in technology)
– Explicit relationships derived from financial models (e.g., structural credit model to relate equities and credit, CDS-Bond spreads) – Conditional scenarios based on simple correlations (expectation or a statistic drawn from their joint distribution, conditional on the outcome of a subset of factors) or more generally use of stressed parameter views – Statistical bootstrapping – Network models (mixture of historical and expert views) – Bayesian methods to combine historical and expert information
– Historical/non-parametric search methods – Advanced data mining tools and self-organized knowledge tools 27
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Historical Crisis Scenarios • Dot-com‌ the plot
Source: S&P Capital IQ - For illustrative purposes only.
28
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios 1. Scenarios are not about the risk factors… they are about portfolios
2. Scenarios tell stories – we need to become much better story-tellers 3. Scenarios must be comprehensive and robust
4. The world is complex – scenarios must represent this world – Continuous development of advanced scenario generation tools – Look into other disciplines like the physical and social sciences
29
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Complex Scenarios Integrating economic and market data, company fundamentals, credit analysis, factor models
Fundamental credit analysis
Market prices
(internal ratings)
PD and LGD models PD f X1 , X 2 , X 3
Pricing model T
PV CF j Y e
r j s j X t j
j 1
• Pricing • Return analysis • Expected losses • Capital • Model risk • Liquidity
Scenarios Factors (issuer, structure, market)
Analytics Integration
Source: S&P Capital IQ Portfolio Risk. For illustrative purposes only.
30
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Complex Scenarios • Example from physics: Sandpile Model • A crisis in the form of a large avalanche is bound to happen – but difficult to reliably predict when • Policy intervention aimed to prevent large avalanches – It should not act on collections of single grains – It should aim at defusing the basic mechanism that underlies self-organization towards criticality
Source: www.math.hmc.edu 28th November 2013 - Abelian Sandpile Model: Symmetric Sandpiles. Natalie J. Durgin. Harvey Mudd College. March 20, 2009. Natalie J. Durgin.
31
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios 1. Scenarios are not about the risk factors… they are about portfolios 2. Scenarios tell stories – we need to become much better story-tellers 3. Scenarios must be comprehensive and robust 4. The world is complex – scenarios must represent this world
5. Decisions must be simple and actionable – The WHAT IF… – Results must be represented and explained in simple terms… at the portfolio level
32
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Example: Historical Crisis Scenarios
Source: S&P Capital IQ Portfolio Risk. For illustrative purposes only.
33
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios – Where’s The Future 1. Scenarios are not about the risk factors… they are about portfolios
2. Scenarios tell stories – we need to become much better story-tellers 3. Scenarios must be comprehensive and robust 4. The world is complex – scenarios must represent this world 5. Decisions must be simple and actionable
6. We cant predict the future… but we have opinions… – Experts, analysts, economists have views which should be incorporated into the risk analysis through scenarios – Advanced tools to integrate data-driven and expert scenarios
34
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
A Story: A Portfolio Manager, A Research Report & Her Portfolio
Source: S&P Capital IQ Platform as of April 03, 2013. For illustrative purposes only. Goldman Sachs: Where to Invest Now, Scaling the Peak. Published February 23, 2013, Accessed April 03, 2013.
35
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Re-Thinking Scenarios 1. Scenarios are not about the risk factors… they are about portfolios
2. Scenarios tell stories – we need to become much better story-tellers 3. Scenarios must be comprehensive and robust 4. The world is complex – scenarios must represent this world 5. Decisions must be simple and actionable 6. We cant predict the future… but we have expert opinions…
7. Scenarios must be shared – scenarios going “social” – Easily accessible and shared across stakeholders Internal: PMs, colleagues, senior management; external – investors, regulators, business and trading partners, external servicers and consultants
– Sharing and communicating scenarios will become more important Scenario libraries & apps, “social scenario networks”, conversations, information sharing
– We need more intelligent risk systems…
36
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Scenarios And Intelligent RISK
Risk systems will be
intelligent Fully interactive and integrated with its users, • Delivering and synthesizing information in real-time, • Modeling and aggregating the risk of multi-asset portfolios in a fully consistent way,
• Modeling the future through rich scenarios which are shared across all stakeholders and social media
37
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Some Advanced Scenario Tools To Watch for... • Rules-based scenarios – normalized and standardized • Automatic scenario generation from investment and research reports, news (NLP) • Reverse stress-testing and inverse methods • Network models and tools: causality networks, Bayesian networks, graph methods • Advanced statistical inference and data mining – historical data where available should inform simulations • Machine-learning methods and predictive analytics
• Self-organized knowledge mining • Agent-based modelling – Social and financial systems are complex adaptive systems with agent interacting at multiple temporal and spatial scales
• Multi-scale modelling – modelling of multiple temporal and spatial scales, from massive and heterogeneous data 38
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
Intelligent And Responsible Use Of Risk Analytics • We need analytics...but we need to understand the application and limitation of our models
• Regardless of the approach, we must understand – The details: the structure, behaviour, and all the risks of individual instruments and portfolios – The limitations of the data and prices, specially when liquidity is limited – Understand “Knightean” uncertainties – The “heroic” assumptions in our models
• Vital to develop model risk approach and comprehensive stress testing
Think, breath, apply, share 39
Scenarios!
Permission to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.
www.spcapitaliq.com Copyright © 2014 by Standard & Poor’s Financial Services LLC (S&P). All rights reserved. No content (including ratings, valuations, credit-related analyses and data, model, software or other application or output therefrom) or any part thereof (Content) may be modified, reverse engineered, reproduced or distributed in any form by any means, or stored in a database or retrieval system, without the prior written permission of Standard & Poor’s Financial Services LLC or its affiliates (collectively, S&P). The Content shall not be used for any unlawful or unauthorized purposes. S&P, its affiliates, and any third-party providers, as well as their directors, officers, shareholders, employees or agents (collectively S&P Parties) do not guarantee the accuracy, completeness, timeliness or availability of the Content. S&P Parties are not responsible for any errors or omissions, regardless of the cause, for the results obtained from the use of the Content, or for the security or maintenance of any data input by the user. The Content is provided on an “as is” basis. S&P PARTIES DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE, FREEDOM FROM BUGS, SOFTWARE ERRORS OR DEFECTS, THAT THE CONTENT’S FUNCTIONING WILL BE UNINTERRUPTED OR THAT THE CONTENT WILL OPERATE WITH ANY SOFTWARE OR HARDWARE CONFIGURATION. In no event shall S&P Parties be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs) in connection with any use of the Content even if advised of the possibility of such damages. Credit-related and other analyses, including ratings, and statements in the Content are statements of opinion as of the date they are expressed and not statements of fact or recommendations to purchase, hold, or sell any securities or to make any investment decisions. S&P assumes no obligation to update the Content following publication in any form or format. The Content should not be relied on and is not a substitute for the skill, judgment and experience of the user, its management, employees, advisors and/or clients when making investment and other business decisions. S&P’s opinions and analyses do not address the suitability of any security. S&P does not act as a fiduciary or an investment advisor except where registered as such. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives.
S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of their respective activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain non–public information received in connection with each analytical process. S&P may receive compensation for its ratings and certain credit-related analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, www.standardandpoors.com (free of charge), and www.ratingsdirect.com and www.globalcreditportal.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at www.standardandpoors.com/usratingsfees. To the extent that regulatory authorities allow a rating agency to acknowledge in one jurisdiction a rating issued in another jurisdiction for certain regulatory purposes, S&P reserves the right to assign, withdraw or suspend such acknowledgement at any time and in its sole discretion. S&P Parties disclaim any duty whatsoever arising out of the assignment, withdrawal or suspension of an acknowledgment as well as any liability for any damage alleged to have been suffered on account thereof. STANDARD & POOR’S, S&P, CAPITAL IQ and GLOBAL CREDIT PORTAL are registered trademarks of Standard & Poor’s Financial Services LLC. S&P CAPITAL IQ is a trademark of Standard & Poor’s 40 Permission Financial Services LLC.to reprint or distribute any content from this presentation requires the prior written approval of S&P Capital IQ. Not for distribution to the public.