Events 2002-2003

This page contains information about events held in the centre in the academic year 2002-3. Many of the talks given have slides available, which can be downloaded by clicking on the pdf icon () next to the talk's title. Some talks also have related papers available for download, which can be accessed by clicking on the paper icon () to the right of the talk's title.

The events held were

Operational Risk, Ruin and Time Change

Risk Measurement and Investment Performance

Credit Risk and the Cost of Capital

Financial Contract Representation

Hughes Hall Lecture

Long Term Investment

Valuation of Asian Options

Credit Risky Fixed Income Instruments

Operational Risk, Ruin and Time Change

Friday, 13 June 2003

Prof. Paul Embrechts

ETH, Zurich and Centennial Professor of Finance at LSE

Ruin, Operational Risk and How Fast Stochastic Processes Mix

Since Basel II, operational risk has been added as a further risk class added to credit and market risk. After a short introduction to the state of the art with respect to Basel II, I will discuss the stylized facts of operational risk data. Based on this, I will discuss methods from the realm of actuarial mathematics which may be used as tools under the so-called Pillar 1 Advanced Measurement Approach. Motivated by the analysis of the above problems, I will present extensions of classical ruin problems in the presence of large claims. The notion of time change, operational time, is important here.

Risk Measurement and Investment Performance

Friday, 9 May 2003

Dr Christian S Pedersen

Mercer Oliver Wyman, London

Selecting a Risk-Adjusted Shareholder Performance Measure

The emergence of 'alternative' investment opportunities, the current bear market and the Wall Street analysts' conflict of interest debacle, have put pressure on current investment performance measurement methodologies. We present a survey of classic and modern performance measures and assess them against objective criteria. Depending upon the market, industry or group of assets studied and the preferences of investors, different measures gain favour and we propose key questions to address when selecting an appropriate performance measure. Our arguments are demonstrated empirically for the global financial services sector, for which we document strong evidence in support of using Sharpe Ratio-based measures. As a comparison, we also look at firms listed on the UK Alternative Investment Market (AIM) for which we illustrate a divergence of rankings based on alternative measures. General implications for risk management and asset allocations across different asset classes are discussed.

Dr William F Shadwick

The Finance Development Centre Limited
(Joint work with Ana Cascon, Con Keating and Bradley Shadwick)

Omega Function Analysis of Financial Returns: Persistence of Performance in UK Investment Trusts

The Omega function of a distribution is a new mathematical tool for analysing the distribution's properties. Like the characteristic function, it is equivalent to the distribution itself and is defined for any distribution which has a mean. Applied to the returns distribution of any financial instrument, it provides a natural measure of relative performance. As it is a global function of the distribution, the local (shape) characteristics of the Omega function contain the effects of fat tails, asymmetry and other properties which are critical for differentiating between the performance of funds, managers or other financial instruments, based on their historic record. Standard mean/variance analysis ignores these effects and extensions based on higher moments are limited to the cases in which these moments exist and may be estimated accurately from sparse data. As the Omega function is defined even for distributions which have no moments higher than the first, it may be applied without assumptions about or estimates of higher moments and has proved effective even with relatively small samples.
We have applied this approach to numerous examples of real investment returns data to address problems of performance measurement, risk analysis and portfolio optimization. Here we report on a study of monthly returns data from four UK investment trusts over the 12 year period ending in January 2003. We have found clear evidence of persistent differences in the returns distributions over the first 10 years which correctly predict the terminal values at the end of the last 2 out of sample years. What is more, the predictions which this analysis made over the two non-overlapping 60 month subsamples and over the 3 non overlapping 40 month subsamples are all the same. As a result, the outcome of the 12 year rankings by terminal value is what could have been predicted at the end of the first 40 months. In this talk I will begin with an introduction to Omega functions and their relevant properties, then describe the analysis of the investment trust data.

Credit Risk and the Cost of Capital

Friday, 25 April 2003

Professor Hashem Pesaran

Faculty of Economics and Politics and Trinity College

Macroeconomic Dynamics and Credit Risk: A Global Perspective

The aim of this paper is to develop a framework for modeling conditional loss distributions through the introduction of risk factor dynamics. Asset value changes of a credit portfolio are linked to a dynamic global macroeconometric model, allowing macro effects to be isolated from idiosyncratic shocks from the perspective of default (and hence loss). Default probabilities are driven primarily by how firms are tied to business cycles, both domestic and foreign, and how business cycles are linked across countries. The model is able to control for firm-specific heterogeneity in an explicitly interdependent global context, as well as to generate multi-period forecasts of the entire loss distribution, conditional on specific macroeconomic scenarios. The approach can be used, for example, to compute the effects of a hypothetical negative equity price shock in South East Asia on the loss distribution of a credit portfolio with global exposures over one or more quarters. Our conditional modeling framework is thus a step towards joint consideration of market and credit risk. The approach has several other features of particular relevance for risk managers, such as the exploration of scale and symmetry of shocks, and the effect of non-normality on credit risk. We show that the effects of such shocks on losses are asymmetric and non-proportional, reflecting the highly non-linear nature of the credit risk model. Non-normal innovations such as Student t generate expected and unexpected losses which increase the fatter the tails of the innovations.

Dr Thomas C. Wilson

Managing Director and Global Head, Finance and Risk Practice, Mercer Oliver Wyman, New York

Issues in Valuing Financial Institutions: The Cost of Capital

● Linking internal performance metrics (RAROC, Economic Capital) and market value metrics (P/E, P/B): The role of capital, cost of capital, performance and growth
● The cost of capital: Theory and Evidence for three important questions
● Is a common hurdle rate appropriate for a financial institution if capital is allocated according to risk?
● How does leverage impact a firm's cost of capital?
● How does a firm's idiosyncratic risk impact its cost of capital?

Financial Contract Representation

14 March, 2003

Professor Michael Dempster

Centre for Financial Research, Judge Business School

Current Alternative Approaches to Formal Contract Representation

Derivative contract trading volumes continue to expand exponentially with the exchange traded instrument business growing and the OTC business slowing in some areas and very rapidly increasing in others such as credit derivatives. All this activity has increased the pressure on major institutions to automate the processing of derivative trades throughout their life cycle and across asset classes. The key to this automation is formal representation of financial contracts. After setting the background, this talk will survey the current available alternatives for derivatives: interbank, standard representations such as FpML, proprietary valuations systems whose emphasis is on efficient pricing and hedging and formal (functional) languages such as Lexifi's MLFi. The emphasis in the talk will be on the operational functionalities supported and not the software details.

Dr Jean-Marc Eber

Lexifi Technologies and INRIA, Paris

Describing, manipulating and pricing financial contracts: The MLFi language

Based on our current implementation of the MLFi language, we give an introductory informal presentation on a formal language for describing financial (bilateral) contracts. We show the feasibility and practical industrial importance but also difficulty, of this approach. We emphasize in particular the benefits of a contract description that is independent of any particular use (in particular of any valuation method!). Our compositional contract "algebra" makes it possible to give an unambiguous precise meaning to any defined contract, a necessary condition for reasoning about contracts. We then show two possible (and implemented) uses of such contract specifications: (1) the automatic generation of pricing programs (given a model description, of course) and (2) the automatic "back-office" contract management through time. We give hints about other possible uses. We show intuitively how these approaches are strongly linked to two concepts well known from Computer Science: denotational and operational semantics. This talk is targeted toward a finance theory audience, therefore no deep knowledge of language theory will be assumed and only simple intuitive examples will be presented.

Hughes Hall Lecture

3 March, 2003

Stephen King

HSBC Investment Bank

After the Technology Bubble

Long Term Investment

28 February 2003

David Buckle

Director of Research, Lee Overlay Partners

Modern Portfolio Theory : Should Active Managers be Using it?

An active manager who manages an investment portfolio on behalf of the portfolio's "trustee" must target the trustee's objective. A trustee makes a single, long run, allocation decision between active managers and the market portfolio, therefore the objective for the active manager is to maximise information ratio. This being the case, the active manager will not maximise information ratio by using modern portfolio theory, namely mean/variance optimization. Instead we show in our theorem of active portfolio management that mean/second-moment optimization. is the optimal investment approach.

Dr Bart Dowling

Director, Global Asset Allocation, Merrill Lynch

Nature, Nurture and Economic Growth

This paper develops an endogenous growth model to examine the influence that heterogeneous entrepreneurial ability between individuals and between countries can have upon economic growth. Entrepreneurial ability is argued to be a function of the combined influence of societal, psychological and genetic factors, the net effect of which upon any individual remains difficult to forecast. Failure by financial markets to accurately identify true entrepreneurial ability ex-ante can result in unfulfilled expectations ex-post. Overestimation of true entrepreneurial ability by imperfectly informed financial institutions causes a divergence between fundamental and market values with the subsequent emergence of a speculative bubble - as consistent with the recent US stock market experience. That said, the news isn't all bad. Even though speculative bubbles may result in a mis-allocation of capital from a rational expectations standpoint, there is still a net benefit of knowledge spillovers generating a higher growth outcome even when over-investment exists. Again, this may be consistent with the experience with the recent bubble in the US market.

Valuation of Asian Options

14 February 2003

Dr Vicky Henderson

Nomura Research Fellow, Nomura Centre for Quantitative Finance, University of Oxford

Bounds for Floating-Strike Asian Options using Symmetry

● Floating vs Fixed Strike Asian Options
● Symmetry between Floating and Fixed Asians for 'forward starting' and 'starting' cases
● Exact solution for Floating Strike as Fixed Strike Asian, utilising existing techniques for Fixed Strike Options
● Upper bound for 'in progress' Floating Strike Asian option
● Bound as combination of Fixed-Strike Asians, vanilla options, optimised over weighting parameter
● Choice of optimal weighting parameter
● Computations of bound

Dr Christoph Burgard

Quantitative Analytics Group, Barclays Capital

Accurate Pricing of Asian Options using PDEs

● Comparison of different methods for pricing Asians
● PDE pricing of discretely sampled Asians
● IN/Out Asians, Asian ratios, Asians with barriers, Cliquets
● Practical hedging strategies for Asians

Credit Risky Fixed Income Instruments

11 October 2002

Dr Luke Olsen

Barclays Capital, London

Pricing Convertible Bonds With Credit Risk

Convertible bonds are a class of corporate debt securities that may be converted at the investor's option into equity. The global market for convertible products has grown rapidly in recent years, as their financing advantages, flexibility and risk profiles have become better understood. This seminar examines the key features of convertibles, their appeal to issuers and investors, traditional valuation methodologies and how their limitations have led to more realistic models. In particular, the relationship between a corporate's credit risk and its share price is crucial for convertibles. We demonstrate how a tractable, extended model to include a single credit-equity sensitivity parameter captures key market behaviour such as 'soft' bond floors, increasing delta and lower (or even negative) volatility sensitivity. This has important implications for risk management using convertibles in the primary and secondary markets.

Dr Alex Taylor

Judge Business School, University of Cambridge

Liquidity and Bond Market Spreads

Recent research by Elton et al (2001) argues that investment-quality defaultable debt spreads reflect three factors: expected losses, risk premiums and taxes. In this paper, we sort bond price data on liquidity proxies (quote frequency, bond age and issue size) and show that an important additional component of spreads is a liquidity premium.