Polynomial Process and Polynomial Regression Model with Interactions of Electricity Prices

Electricity is vital to our modern life. Different from other commodities, electricity prices show seasonal patterns with mean reversion and ‘short-lived’ spikes. It is significantly important for market participants to model power prices (both spot prices and forward prices) for risk management. Polynomial process is a class of time-homogenous Markov processes with the property that expectations of polynomial functions (of degree n, say) of the future state of the process, conditional on the current are given by another polynomials (of degree <=n) of the current state. Once we could model the observed factors such as market demand and unit costs of generating power by different energy commodities with polynomial processes, then we could apply polynomial regression to model the spot prices of electricity and this great property enables us to find the explicit formula for forward prices. Finally we expect to conveniently capture the dynamics of electricity market and price the related derivatives.

Faculty Supervisor:

Antony Ware

Student:

Partner:

Université Paris Dauphine

Discipline:

Mathematics

Sector:

Finance and Insurance; Energy and Utilities; Oil and Gas

University:

University of Calgary

Program: