Exchange Ideas
Defence of the Rocket ScientistThis blog is dedicated to quantitative risk management and aims to provide an enhanced perspective on the techniques borrowed by finance from science and engineering May 13, 2010 Dealing with the limits of scienceEver since the crisis began - some will say way earlier than that - there've been those who said that risk can never be measured and as such, all efforts in that direction are a waste of time and a distraction from managing the business. The crisis added a new dimension to the debate, by accusing the rocket scientists of being a cause of the crisis. So this works like this: I'm going to a tailor and order a pair of flary trousers to impress my girlfriend (don't comment on the tastes), she ridicules me for the bell-bottoms, then I complain that the trousers are too large at the ankles because the tailor's scissors weren't sharp enough or that he wasn't careful when he operated the scissors. The RiskMetrics Technical Document, arguably the blueprint of the widest used - and most maligned - measure of market risk, VaR, makes clear from the beginning (Chapter 2, "Historical perspective of VaR") the limitations of the statistical approach proposed within. The caveats are not, as may seem at first read, simply limitations of liability; in fact, at the top of page 22, the document recommends that "Risk managers should use both approaches — the statistical approach to monitor risks continuously in all risk-taking units and the scenario approach on a case-by-case Now, any rocket scientist starting to read the document may be forgiven for skipping the above recommendation - their brief was not, until very recently, to address scenario analysis but to produce more and more sophisticated versions of the statistical approach. Of course, any scientist worth her salt needs to state clearly that when data are sparse, statistics will be shaky and potential for financial loss greater than estimated - but we assume that scientific integrity survives the transition to the financial world. So what should a rocket scientist do when asked to model something she shouldn't, like firm-wide operational unexpected loss at 99.9% confidence? a. Do it at 75% and state that this is the best that current science can achieve? b. Do it at 99.9% and ignore quantile estimation errors, in the safe knowledge that few other scientists and nobody else can spot the crucial omission? c. Reject the instruction, stating that the work cannot be done as requested, and change jobs? The limits to our science and technology are painfully obvious all around us - space, oil fields, medicine, etc - so, when snake oil is offered for sale, the buyer better beware - whether that is a trader 'buying' the model from the quant or, more often than not, the trader's client buying overpriced or 'faulty' financial contracts. The failure of quantitative finance damns us all, buyers and vendors alike, but rocket scientists bear less responsibility than most for it all. Posted by Dan at 05:00 PM | Comments (0) October 19, 2009 Risks in the CloudCloud computing has made the front page of The Economist last week, after years of steady progress. It started out as a revolutionary way of performing everyday computing tasks, and now the big firms are battling for supremacy in this area – and indeed, for the future of computing, many billions of revenue for years to come, and for some, perhaps their very survival. Everything will be thrown into battle, from the most modern viral marketing to classic legal action; the consumer will just have to stand back and wait for the winner to emerge… On the face of it, who could argue with the cost advantages and the convenience of cloud computing? Desktop computing – a manufactured commodity product – is being replaced by services delivered by a utility. But let’s examine the risks: 1. the services require something that the customer owns: data. How is cloud computing protecting data ownership and the value to their owner? Sensitive information will still have to stay ‘in-house’ throughout the entire workflow. A version of this risk is particularly worrying – the State can invoke the national interest, and information which in the past had to be gathered is now all in one place, ready for analysis and manipulation. 2. one of the strengths of the Internet was its built-in redundancy. But in an ideal world (for the giant providers of cloud computing), a small number of companies will share the market between them; should any of them fail for any reason, large numbers of customers will be left without even the basic computing capabilities. The comparison with utilities is particularly apt here, remembering the massive blackouts which happened in recent history on the East Coast of the US and Canada. 3. recent cyber attacks have exposed vulnerabilities unknown before; as computing power is concentrated in the massive data centres that the big players are building, it will become so much easier to knock out large parts of the economy. The list can go on, but let’s see if we can quantify the risk, and whether the savings are worth the price of risk. In one scenario, what is the financial impact of a total computing blackout of a bank, lasting 8 hours? It can be modelled with one of the accepted stochastic distributions for loss severity, like the lognormal distribution, and for the sake of this argument let’s say that the expected loss would be 1 day profits . The probability of impact is low, to be sure, hard to estimate from historic data; what if we take it as a one in 10 year event? The resulting loss would be 1 day profits in 10 years of operation. Is it worth all the savings brought by The Cloud? Posted by Dan at 09:47 PM | Comments (0) August 21, 2009 About Dan OprescuDan Oprescu is Managing Director of Operational Risk Analytics Ltd (www.operationalriskanalytics.com), a business that helps banks and insurers comply with the new, risk-based regulation of capital adequacy. For many years, Dan has been a Visiting Fellow at Cass Business School, London, where he teaches Risk Analysis and Modelling for the M.Sc. in Mathematical Finance. Dan has more than 15 years in Financial Mathematics and Risk. He was formerly as a consultant in Trading and Risk at Sungard; prior to that, he worked in quantitative research and development with Barclays Global Investors. Dan earned a PhD in physical chemistry from University College London, with a thesis on the Hilbert transform between electronic and molecular spectra. He also has an MSc in Mathematical Finance from Cass Business School, London. Dan has published his scientific research in academic journals; he continues to publish in financial practitioner journals, and speaks regularly at conferences. Dan's main interests are Risk Modelling, Risk Aggregation, Economic Capital, and Regulatory Risk and Capital (Basel 2 and Solvency 2). Posted by Dan at 10:00 AM | Comments (0) |
 CategoriesArchivesRecent Entries |
