The purpose of this paper is to model the stochastic behavior of the nodal prices of electricity in deregulated markets in the USA, and in particular, to explain how this behavior has changed over time. These changes require different modeling strategies to capture the salient features of the market. The paper discusses three different models. The first model represents the period immediately after deregulation when price spikes were observed during the summer months using stochastic regime switching. Regulators in the northeastern markets responded to these "uncompetitive" price spikes by implementing various forms of market monitoring to mitigate the speculative behavior of generators. However, in a market like New York State, congestion on the transmission lines transferring power to New York City still led to spatial price differences. A new market for Transmission Congestion Credits (TCC) was established to allow generators to hedge these price differences. The second strategy uses Vector Autoregressive (VAR) Models of temperature, demand and price to estimate the financial riskiness of holding a TCC. Finally, current research on modeling the demand for electricity on networks with high penetrations of renewable sources of generation is discussed. Since these sources of generation are generally intermittent, it is likely that controllable demand, such as charging electric vehicles and thermal storage, will be increasingly important as a way to mitigate the variability of supply. These technologies present new challenges for modeling the demand for electricity. It will be important to capture the ability to shift demand from peak to off-peak periods as well as to mitigate the variability of renewable generation.
展开▼
