III. A GENERAL POWER CONTRACT PORTFOLIO OPTIMIZATION PROBLEM
In a deregulated power market, individual generation company strives to optimize its portfolio to achieve profit maximization. At the same time, managing a power portfolio becomes a risky business because deregulation brings in market uncertainty as well. In order to hedge against market risks, various power derivative contracts have been developed, such as swaps and swaptions which are discussed in the previous section. Although the optimization objective remains unchanged, the innovation of supply instruments should be accommodated in portfolio components, and more sophisticated risk measures should be catered.
With regard to risk measures, variance is incapable of dealing with the asymmetrical profit/loss distribution of a power portfolio as recognized among researchers. VaR is a breakthrough to aggregate risk across an institution, which summarizes the worst loss over a target horizon that will not be exceeded with a given level of confidence [15]. However, VaR suffers the criticism that it is not subadditive. A coherent risk measure CVaR which is derived from VaR is adopted here to form the risk constraint.
We argue the horizon for a single period optimization problem should be chosen as one month. This is because monthly swaps and their corresponding swaptions are widely traded, and it is more critical for the generation company to consider monthly rather than yearly profit/loss in order to reserve enough capital to cover the exposed intermediate risks.
The case here is a power portfolio considered by a generation company, and this portfolio consists of various power supply contracts. The loss function of the portfolio is
defined as f (O,y) = —cOy, where COis the decision vector
The most widely traded options at AEM are not swaptions but caps.
đang được dịch, vui lòng đợi..