Exchange Ideas
A Quant's eye view of Risk ModelingIt focuses on the contemporary issues in risk modeling faced by Practitioners and Researchers and explores possible roadmap to resolve it December 04, 2010 Credit Crisis and the GreyING Anatomy: Touring from Germany to Romania in presentation of my thoughts in risk conferencesCredit Crisis is not an event now; this has grown into a starting point of a series of systematic risk-inducing financial events that may trigger further crisis. I am not bullish about regulatory measures like proposed Basel III , as I believe these measures of forcing the banks to push up higher capital asset ratio and shrink risk assets will lead to plunge on M3 ( largest in Money Supply) which was never successful historically even during Great Depression. Further my hindsight goes that the self-induced innovations to save on equity may lead further re-engineering which will undermine the basic purpose of capital regulation, to the detriment of the regulators, and instead of pre-empting the crisis, on the contrary, this will fine-tune the proliferate more systematically. While I have focused my present research interest on the effect of post credit crisis and systematic risk, which is in addition to my normal professional life as a risk practitioner, and presented some of my thoughts in risk conferences in Germany and Romania, for feed back to learn further and enhance my research, but still at the end of the visit, I am unable to convince myself that shadow of credit crisis is over yet, at least systematic risk point of view, regulation wise. This is for me a Real Identity Crisis! My presentation in Germany and Aftermath: the other Octobor Fest Indeed the CEQURA conference is more scintillating for me rather than the October fest in the background as the speakers are knowledgeable and the list includes European Central Bank. I was very curious about the Germany as a strong EU country, specially I noted with surprise the comments of European Central Bank President Jean-Claude Trichet that "Their (Basel III) contribution to long-term financial stability and growth will be substantial." You doubt or I doubt! The CEQURA Conference on Advances in Financial and Insurance Risk Management was organized in collaboration with the Federal Reserve Bank of Richmond and the Bayerisches Finanz Zentrum, as a platform for presenting and discussing current developments and fosters the exchange between academics and practitioners from the risk management community. The web link of the conference is My presentation was “Turbulent Times & EVT Approach: Indian Experience” which was based on our paper. This paper is co-authored by me with Professor Dr. Basabi Bhattacharya who was my guide for my completed PhD. We tried to estimate the transmitted effect of credit crisis in Indian stock market applying Extreme Value Theory. Our study showed the evidence of transmitted effect of credit crisis in stock market though many hard-nosed say that our system is too strong to be penetrated, which is too hard for me to believe. However out study is under further enhancement and are waiting for more convincing results. A base version of my presentation may be had at the following link: In fact our understanding, as in our paper, synchronizes with the comments by the keynote speaker Paul Embrechts, Department of Mathematics, Director of RiskLab, ETH Zurich who quoted in his presentation "The Financial Crisis: Warnings, Guilt and a Mathematical Theorem" : I strongly feel thagt enough work is required to address this tail risk as the present Basel III framework does not address the possible non-disclosure of the risk in the tails by present Value-at-risk models and even much-talked-about Gaussian Copula which could do even worse by fattening the tails. I believe there is enough scope to address tail risk through appropriate regulations “Now or Never”, as there is not enough time left. While I am impressed by Cristina Danciulescu of Trondheim Business School on “Backtesting Value-at-Risk Models:A Multivariate Approach” for uniqueness of the approach, Tina Yener of CEQURA spoke on “Operational Risk:Estimation and Effects of Dependencies”, which really highly technically sound addressing operational risk modelling issues using my favourite EVT approach. Kerstin Bernoth of DIW Berlin on “Forecasting the fragility of the banking and insurance sector” is a cse in point and she agreed that “Credit crunch of 2007/08 demonstrated that financial institutions cannot be regarded as self-standing and independent”, which make me understand the Berlin Wall! My presentation in Romania and Aftermath: Looking for the other Dracula Though I could not find my all-time favorite Count Dracula in nice windy breezes of Romania, the other Dracula of credit crisis haunted me in the well-managed conference organized by PRMIA Bucharest. I should thank Andrea and Valentin, the co-regional directors of PRMIA Bucharest Chapter. This is a great learning for me as all speakers are hard core practitioners like me and shared their concerns and remedies. The name of conference was Annual Global Credit Risk Event as organized by Bucharest Chapter of Professional Risk management International Association (PRMIA) in Bucharest in Romania from 11th to 12th November 2010. The speakers include from Moody’s, Reserve Bank of Boston. , KPMG Romania & Greece. The conference web link is My presentation was “The impact of the credit crisis on acceptance and design of credit rating models for banks: the way forward" where I tried to estimate the effect of the credit crisis on design and validation of internal rating model as the challenges of discriminatory and calibrating power of the rating systems take a paradigm shift during and post crisis. The dilemma haunts me whether Through-The-Cycle (TTC) rating system need be supplemented by a Point-In-Time (PIT) rating system. A basic version of my presentation may be had at the following link: As I see the challenge post crisis: While the logic of the risk assessment process along with the rating system’s design and operation need to be revisited now, what is most important is to establish the right feed back loop post credit crisis. In my presenation through discussion, I tried to explore what could a preferred road map to address these issues, as the audience are seasoned bankers/practitioners. The IRB approach which had a great touch point with my presentation, I am really impressed by the progress made by Romanian banks, even under many constraints. The presentation by Ada Valcea, Executive Director, Group Risk Controlling & Portfolio Management , Raiffeisen Bank S.A. Romania (One of the largest Banks in the region) addressed issues under the regulatory framework, which are : As I see it , there is reaslistic requirement to revisit the choice of models, existing or prospective – as a preferred choice in the current context. among the following, based on issues of localization and benchmarking: While I was vey moved intellectually by the presentation of Angela Manolache, Director, Financial Services, KPMG, Romania , on “Proposed changes to the IFRS impairment model for financial assets (IFRS 9)” which is amazingly detailed ( too detailed for me!) and content-rich, I am rather confused about credit stress testing method proposed by IFRS regime which is significantly different by Credit stress testing method prescribed by Basel II and will give two different results, adequate enough to confuse senior management. I doubt why no synchronization is made between two regulations where end objective of the bothe regime is to achieve solvency of the bank. Andrea Dochia , Senior Manager, KPMG, Romania, spoke on Credit risk stress testing in the ICAAP (NBR R18/2009). Challanges for banks using the standardized approach for credit risk” which is very content-rich.The key concern I still have about credit stress testing is about factoring in Macroeconomic variables like the effects on credit risk (measured by NPL) of real GDP growth and credit volume growth, as well as the effects of unemployment rate and lagged inflation rate, where regulators are not prescriptive, discreetly! However I am impressed by Preston Thompson , AVP, Federal Reserve Bank of Boston on his talk on “Aligning Risk, Risk Appetite and Capital” which is in fact I am trying to implement professionally, as I see this is the day for identifying adequate headroom for economic capital in compliance with risk appetite. No adequate headroom means no head at all! Afterthoughts Credit Crisis is the risk management marketers’ all-time-favorite dream but a realistic agony for serious risk managers due to sincere doubts about relevance and efectiveness of evolving risk regulations, as most serious risks are SYSTEMATIC in nature and needs to be handled adequately at Macro level rather just at micro level by risk managers.Please send me your comments which make feel I am not lonely in my thoughts! Posted by Dr Debashis Dutta at 12:25 PM | Comments (0) October 22, 2010 About Dr Debashis DuttaDr. Debashis Dutta holds a PH.D. with the dissertation on "Risk Management in Indian Financial Markets" from Jadavpur University,India. His work includes Basel II & specially empirical works on different models of Value at Risk, Back testing, Validation of Credit Rating Model, and Peaks over Threshold (POT) Model of Extreme Value Theory(EVT), used mostly with MATLAB software. He is also an MBA (Finance) from Jadavpur University,India, with project paper on Market Risk Management. Presently he is working with Qatar National Bank as Manager, Group Portfolio Management, Group Risk (Disclaimer: The views and opinions expressed in the blog are those of the author and do not necessarily represent the views and opinions of Qatar Natiional Bank. The information contained is of a general nature and is not intended to address the circumstances of any particular individual or entity).Previously he worked with KPMG, Bahrain & Qatar as Manager, Financial Risk Management practice On his consulting assignments, he has consulted internationally active banks in GCC regions for Basel II implementation/Pillar 2 review and validation of VaR models for internal model approach of Basel II. He has also consulted internationally active banks banks in Jordan for Basel II implementation in his assignment with his previous employer, another consulting company. His recent presentation includes ones made in CEQURA Confernce, Munich, Germany, GES PRMIA Seminar in Bucharest, Romania and in MASR conference, St Petersberg, Russia, on various risk related issues mostly concentrating on model issues post credit crisis. His publication includes "Extremal Index and clustering in the Extreme Values: A study on NSE CNX Nifty"(co-author Dr Basabi Bhattacharya), published in the ICFAI Journal of Applied Finance, issue September 2006 . He has also presented papers at various seminars like "A Bootstrapped Historical Simulation Value at Risk Approach to S&P CNX Nifty" (co-author Dr Basabi Bhattacharya) at National Conference on Money and Banking, Indira Gandhi Institute of Development Research(IGIDR), held at Mumbai,India,2008, "Modelling Extremal Behaviour of S&P CNX NIFTY : An EVT approach"(co-author Dr Basabi Bhattacharya) at National Conference on Money and Banking by Indira Gandhi Institute of Development Research(IGIDR), held at Mumbai,India,2005, "Value at Risk: A study on Indian Financial Market" (co-author Dr Basabi Bhattacharya)at National Conference on Finance and Economics, ICFAI Business School, held at Bangalore, India, 2004, and "Managing Market Risk in the India Context - An assessment of Alternative VaR methods" (co-author Dr Basabi Bhattacharya) as Working Paper Series, Department of Economics; Jadavpur University, Kolkata, India. His strength is risk modeling as he believes in being a Quant is a to quantify and manage the Omnipresent Risk Demon in financial world, untill the God turn the Dice:) Posted by Dr Debashis Dutta at 12:20 AM | Comments (0) November 22, 2009 Macro StressTesting : The King has no clothesStress Testing is not a new agenda now. Rather it is a popular agenda with improper discussion and undue coverage on insignificant results. World Premier of “US Stress Test” attracted much lime light. It unduly focused on insufficient selection of two scenerios, only extending to 2010 and injudicious selection of only three macroeconomic factors. The obvious choice should have a full Monte Caro simulation of all risk factors for 20 years. This raises doubt about FSAP program’s objective, which unfortunately resembles a marketing Quick Win strategy. Further, accounting projections are required to be supplemented by MtM, which was never thought of. This shows that several black holes are plugging other black holes, which is risky and dangerous. It is a matter of fact, post traumatic learning from the “financial tsunami” of the subprime mortgage crisis, the stress Testing has evolved as an integral part of the overall risk governance and risk management culture of the bank. Present credit crisis has necessitated taking liquidity risk into stress testing framework in addition to credit risk and others. The manifestation of liquidity risk can rapidly move the system into tail of loss distribution through bank runs. Further macroeconomic consequences of rating sensitive capital have proved that a procyclical consequence of minimum capital requirement is the order of the day. Failure of efficacy of many models in credit crisis has driven the agenda that stress test should complement complex risk management models. The outcome of the forward-looking stress testing will be dynamic in nature and would guide the bank how the plausible events would adversely impact the banks. This also gives rise to a comprehensive stress testing framework under Pillar II, which encompasses all risks, and this will be subjected to supervision of Central Banks and be incorporated in ICAAP framework. There is a migration from conventional micro models to macro models due to economic downturn. Specially there have long been dependence on macro economic variables like dependence of PDs (Probability of Default) on macroeconomic variables and dependence of exposure at default and collaterals on variables like interest rate which are partially macro. Cobweb of Regulation : Banking on Basel The Stress Testing is viewed by regulators as a key component of supervisory assessment process to identify the vulnerabilities and evaluate bank’s capital adequacy. However, the Stress Testing for Market risk a old phenomenon which is well elaborated in the book elsewhere. Later growing emphasis is given to Credit Stress Testing factoring the macroeconomic scenarios and later focus is on Pillar II stress Testing which takes into consideration of all risks including reputational risk. The Basel Committee on Banking Supervision has issued Principles for sound stress testing practices and supervision on May 2009 (Final Version) which put forth a comprehensive set of principles as guidance for the Banks. The guidance sets for weaknesses that affected the performance of stress testing during the crisis which are broadly classified as follows: • Use of stress testing and integration in risk governance Under use of stress testing and integration in risk governance, BIS emphasized the appropriate use of stress testing in banks’ risk governance and capital planning. The business areas used to believe that the stress testing is not credible. This requires to be changed and an integrated framework rather than silo wise approach would be implemented. This would pave the way for an overreaching approach rather than mechanical approach. Under stress testing methodologies, BIS emphasized the inclusion of the strong interlinkages like for example that exists between lack of market liquidity and funding liquidity pressure. There is a range of practices in stress testing from the simple sensitivity to complex stress testing. This also takes varying degrees of aggregation e.g. from a level of an individual instrument up to the institutional level. Though many cases, the historical relationship was used, the credit crisis has proved to be inefficient by underestimating the interaction between risks and the firm-wide impact of severe stress scenarios. Under scenario selection, BIS emphasized on sophisticated scenario selection, which enables shocks to many parameters simultaneously. While scenarios could be either historical or hypothetical, the weakness found in historical ones in credit crisis is that it were not able to capture risks in new products which have driven the crisis. Furthermore, • the behaviour of complex structured products under stressed liquidity conditions
In the midst of the Birth Pains: Macroeconomic Stress Testing Model A macroeconomic stress testing model generates projections of macro variables as deviations from a base line scenario. The framework for macro stress testing will require the selection of extreme but plausible shocks. These can be univariate shocks in single risk factors or multivariate scenarios with various (macro) risk factors change. For example we may use a depreciation of the dollar exchange rate is combined with a falling GDP and rising interest rates. The scenarios can be developed through the stochastic simulations of macro variables. (i) a structural econometric model The structural macroeconomic models are used to project the levels of key macroeconomic indicators under stress circumstances, where shocks are as exogenous inputs. For FSAP (i.e. ) exercise, the authorities applies the domestic macroeconomic models developed for monetary policy purposes. Vector Autoregression (VAR) or Vector Error Correction Model (VECM). A VAR model comprise of GDP, inflation rate, bank loans outstanding, effective exchange rate, and the overnight call rate. Another variant a Global Vector Autoregressive (GVAR) is modeled on country or region specific VECM. The VEM may be used to forecast the stressed EDFs. This model allows the joint shock of macroeconomic variables, which enables vector process to process the stress impact on the variables. A pure statistical model may allow macroeconomic and financial variables to be modeled through a multivariate t-copula. Credit Stress Testing – A Practitioner’s Choice Process 0.1 – Identifying a Stressed Case
Dale Gray and James P. Walsh in paper “Factor Model for Stress-Testing with a Contingent Claims Model of the Chilean Banking System “ (April 2008 IMF Working Paper No. 08/89) proposed risk indicators for the major Chilean banks based on contingent claims analysis, an extension of Black-Scholes-Merton option-pricing theory. These risk indicators are clearly tied to macroeconomic and financial developments in Chile and outside, but bank responses are highly heterogeneous. To reduce the number of variables linked to the banks' risk to a tractable number, they applied principal component analysis. Vector autoregressions of risk indicators with the most significant factors showed strong ties from financial markets and regional developments. The stress test was not stressful enough For Macroeconomic Stress Testing, one of the keys will be how Bank conducts stress tests determining how future minimum requirements could potentially increase under Internal Ratings Based Approach under Basel II. My worries how stressful will be these stress tests based on specific deterministic macro scenarios prescribed by the regulators.This remains an open question, unclothed. Posted by Dr Debashis Dutta at 12:30 AM | Comments (0) January 18, 2009 Reverse Stress Testing: What's the Great Reversal?FSA's recent Consultative Paper 08/24 proposes to introduce Reverse Stress Test requirement. It also proposes to change existing requirements on Pillar II Stress Testing, as well Pillar I requirement for those banks adopting IRB approaches. Reason is obvious: Reverse Sweep the spinning Credit Crisis. The primary objectives behind this new-born Reverse Stress Test are (a) Use high impact stress events which may lead the banks to fail, and So it is a high level approach towards a Quasi Zero Failure Strategy. From modeling perspective, Reverse Stress Test lays bare two points for provocative thoughts: 1. How to measure the mode and magnitude of risk transmission and contagion design under extreme stress, and The Central Banks are not prescriptive; so the open question before the Risk Managers - "HOW?" The ICAAP framework is now being foisted on the banks hurriedly by the Regulatory Authorities in many countries, as part of Basel Pillar II compliance (or Bank Failures?). It is required to capture the "cross-market and cross-risk type dependencies on the assets of the firm as a whole". So, the complete suite for Strategic Risk Management may include the following: 1. Capital allocation "Principles for sound stress testing practices and supervision" Issued for comment on January 2009 by Basel Committee on Banking Supervision gives guidance for stress testing during the crisis and says: "First, given a long period of stability, backward-looking historical information indicated benign conditions so that these models did not pick up the possibility of severe shocks nor the build up of vulnerabilities within the system. Historical statistical relationships, such as correlations,proved to be unreliable once actual events started to unfold." Point noted : Inefficacy of histrorical statistical measures - Break down in correlation pattern. "Second, the financial crisis has again shown that, especially in stressed conditions, risk characteristics can change rapidly as reactions by market participants within the system can induce feedback effects and lead to system-wide interactions. These effects can dramatically amplify initial shocks as recent events have illustrated." Point noted : Risk tramission and co-dependence structure - Contagion effects.
Simplistically speaking, the steps for an ideal stress testing framework may contain the following: 1. Identify risk factors i.e. macro-economic and micro-economic factors which may affect the assets From Quant's point of view, there are two enticing words in the professed Reverse Stress Test: 1. Correlations Generally correlation is used to describe dependence between random variables. But eventually copula has proved its superiority to model dependence, especially in financial risk management. Ideally the dependence model should capture the risk expressed by the joint tail behavior of returns, without compromising the diversification possibilities that may be represented by the center of the return distribution. There are many types of copulas that are used in risk management. Longin (1996) and Jansen et al. (2000) applied extreme value copula. Longin and Solnik (2001), & Poon et al. (2004) opted for the gumbel copula. Glasserman et al. (2002), Campbell et al. (2003), Mashal et al. (2003), Valdez and Chernih (2003), & Meneguzzo and Vecchiato (2004) used t-copula. The process to evaluate the fit of a copula may broadly contain following steps: 1. Estimation - Here we estimate the parameters.
Posted by Dr Debashis Dutta at 10:32 PM | Comments (0) August 07, 2008 POTholding the movement in the tails of financial return series: The story of Peaks Over Threshold (POT) MethodUnraveling the volatilities in the financial markets has always been an undecipherable mystery like intrigue smile of Monalisa. When conventional VaR methods fail to decipher the heightened movement in the tails of the financial returns series, one can do it: The Extreme Value Theory (EVT). While Basel II stipulates for VaR based capital charge under Internal Model Approach, it is important to make the estimate accurate or else may lead to sub-optimal allocation. EVT-based VaR could be a good candidate. The Extreme Value Theory is a branch of statistics, dealing with extreme deviation from the median of probability distribution. There are two classes of theorem based on data generation process: Theorem I : Theorem II : In POT, the excess distribution is taken. It is also called Generalized Pareto Distribution(GPD). The GPD takes three parameters viz. tail parameter, location parameter, and location parameter. The equation of GPD is The challenge arises when the distribution is non-linear. The non-linearity could be ascertained by using a graphical exploration tool like QQ Plot. A QQ plot shall look like this for non-linear data i.e. tending towards straight line. The sample mean excess plot could be used. Next step is to find a threshold and GPD is fitted to the exceedences above the threshold. Then VaR can be estimated by POT-ML estimation for upper bound, point estimation and lower bound. The EVIM package of MATLAB is very useful input for EVT analysis though I use another version , modified by me. EVIM can be downloaded from link (http://www.bilkent.edu.tr/~faruk/evim.htm) i.e. Faruk's webpage along with an amazing documentation by them.
Only I believe Basel must direct the banks to POThold the movement in the tails of the returns series to compute a prudent estimate of VaR and subsequent correct computation of regulatory and economic capital.We can not afford another subprime crisis. Posted by Dr Debashis Dutta at 04:31 PM | Comments (0) June 13, 2008 Basel II, Less Data Points & Bootstrapped Simulation VaRI believe the dimensionality of the data points for VaR computation is really a challenge for Practitioners around the world in Basel II regime. While Historical Simulation Value at Risk method takes care of the volatility and correlation present in original data set, still it requires 3 to 5 years of data or more. Many banks do not have such adequate data points. I strongly believe the Bootstrapped Historical Simulation VaR is a good candidate to address this issue. Continue reading "Basel II, Less Data Points & Bootstrapped Simulation VaR" Posted by Dr Debashis Dutta at 01:26 PM | Comments (0) |
 Categories
ArchivesRecent Entries
|
