The ongoing deregulation of electricity markets worldwide has a major impact on the power industry. New price risks require new risk management tools and new methods for the valuation of generation and transmission assets as well as existing (physical) electricity contracts. As far as risk management is concerned, many derivative instruments have been designed to hedge against spot price risk or different types of liability risk exposure. For instance one observes the emergence of markets for simple financial and physical derivative contracts such as futures, forwards, call and put options, etc. In addition, there is an immense variety of derivative contracts with American style and path dependent payoff structure. Such options are much more widespread in the energy business then in finance. Probably the most important examples are swing options which have been traded in electricity over-the-counter markets for a long time. Swing options are sometimes referred to as virtual power plants. In fact the problem of finding an optimal exercise strategy for a swing option is basically equivalent to the problem of finding an optimal generation schedule for a real power plant. The well known methods of finance (such as stochastic processes, option theory, and stochastic dynamic programming) can essentially be applied to forecasting, scheduling and pricing problems in the energy business, as well. However the special peculiarities of electric power lead to complications. In fact electric energy is not storable. Thus, contingent claims must be hedged by trading in forward contracts and a risk free asset, whereas the spot price of electric energy has to be considered as a non tradable state variable driving the market. Moreover electricity prices are mean-reverting and exhibit jumps and spikes, which significantly complicates the valuation of European-style derivatives. Finally, the most wide-spread derivative contracts (as well as the physical generation assets) in the energy business have an American-style and path-dependent payoff structure. The valuation of such contracts requires solution of involved stochastic programming problems.