Hedging Strategy and Electricity Contract Engineering - IFOR
Hedging Strategy and Electricity Contract Engineering - IFOR
Hedging Strategy and Electricity Contract Engineering - IFOR
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
3.4 Traditional risk management models 51<br />
management models can however give some insight into the challenges that the<br />
electricity market is facing.<br />
3.4.1. Risk measures<br />
Whereas deterministic problems are characterized by ’real’ numbers, such as<br />
costs or returns, stochastic problems are characterized by r<strong>and</strong>om variables.<br />
Naturally it is this stochastic variability that adds the risk component to the<br />
problem. We know that risk should be managed, but a natural question is<br />
how this risk should be measured. The simile taken from Gumerlock & Litterman<br />
[GL98] exemplifies the problem of how to measure risk:<br />
Leaving aside risk for a moment, consider the measurement of human<br />
size. Everyone knows qualitatively what large <strong>and</strong> small mean,<br />
but life gets more difficult when we want to express size in a single<br />
number. Either height or weight can be useful, depending on<br />
the problem being addressed. Each metric is appropriate for a given<br />
problem, <strong>and</strong> neither serves all purposes. Indeed, if one asks for a<br />
definite answer to the question of which metric is the best measure<br />
of size, the answer is that neither height, nor weight, nor a linear<br />
combination of them is the best measure of size. The best measure<br />
of size is the one most appropriate to the purpose for which it is<br />
intended.<br />
As with human size, the same problem arises when measuring risk. Actually<br />
risk is too complex to characterize with one number. On the other h<strong>and</strong>,<br />
management calls for a simple measure of risk, which should be easy to<br />
interpret. One single number has therefore traditionally characterized risk.<br />
A typical risk measure developed for this purpose is Value at Risk. Another<br />
type of risk measure is variance, which however does not possess a natural<br />
interpretation from a management point of view. On the other h<strong>and</strong>, it has<br />
been derived from the utility maximizing agent theory as the risk measure that<br />
would concern such an agent, of course under certain assumptions.<br />
We will here present the traditional risk measure, variance, the quantile based<br />
Value at Risk <strong>and</strong> the blunt stress test to give an idea of the variety of risk