Uncertainty and risk are the mainstays of Alexander Schied’s research.
According to classical financial engineering theory, all risk can be eliminated by using dynamic trading strategies, but in reality such hedging can’t remove risks such as a natural disaster, so financial engineers look for the best way to measure these intrinsic risks. Using these risk measures, Schied develops methods to predict the likely effect on a portfolio.
Having a good idea how much it might lose allows a bank to keep enough money on hand to cover its debts in a worst-case scenario. “The number is usually interpreted as a capital requirement,” says Schied. “A financial institution must hold enough capital to account for the possible shortfalls.”
While financial models can be useful, they are far from perfect, so Schied also studies how to deal with the inherent uncertainty of working with simplified models of impossibly complex reality.
“From a practical point of view it’s very important because people often believe too much in the models they are using. To some extent, this is what happened in the subprime mortgage crisis,” he says. “They have used the wrong models and found the wrong prices and now there is a problem. To some extent it has to do with believing too firmly in their own models.”
By quantifying the strengths and weaknesses of models, Schied hopes to inform investors when they can have confidence in the models and when they need to be more cautious. “This is nothing that has ever been implemented and it’s also quite new,” he says.
A third area of research for Schied, begun when he was scientific director of the Deutsche Bank Quantitative Products Laboratory in Berlin, is liquidity risk. That is what a financial institution faces when it has to sell a lot of a particular holding, as recently happened when Societe Generale had to unload the purchases of rogue trader Jerome Kerviel.
“Immediately, what happens is the price of the underlying stock drops dramatically and you get only a fraction of the face value,” says Schied. “So to reduce your liquidity risk what you have to try to do is split up a large order into many small orders and spread them out over time, but then you get additional volatility risk because the stock price moves.”
Schied’s work in this area helps traders balance these two competing risks to minimize losses. “We have a new model, some new strategies, and some qualitative results about such strategies,” says Schied, “so we can explain some behavior or patterns in such situations by referring to the risk aversion of the agent.”
Prof Schied's Web site
