Estimating Spot Price and Smooth Forward Curve in Electricity Markets with Bayesian Penalized Spline
Date
2014-05-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The first part of this thesis presents a Bayesian penalized spline approach to constructing smooth forward curves in electricity markets. Since electricity should be delivered as a continuous
flow, power contracts have settlement periods rather than a fixed delivery
time. In addition, electricity forward curves have strong seasonal shape. Our approach provides a method for estimating a continuous forward price curve from market forward prices quoted over a period. The approach is illustrated using observed market data from the Mid-Columbia (Mid-C) and California Oregon Border (COB) pricing hubs.
Since Mid-C is a liquid market where forward contracts are quoted every day and COB
is an illiquid hub, a two step estimation procedure is developed from Bayesian perspective.First, the Mid-C smooth curve is constructed using Bayesian penalized spline. Next, the COB smooth curve is estimated by adding a spread to the constructed Mid-C smooth curve and incorporating the positive spread between the two hubs as an informative prior.
In the second part, we present a mean reverting model for the electricity spot price and we employ a Bayesian penalized spline approach to model the deterministic seasonal function exhibited in the monthly averages. Based on the historical spot price from Alberta power market, we calibrated the model parameters, also by implementing Bayesian estimation techniques.
Description
Keywords
Economics--Finance, Mathematics, Statistics, Energy
Citation
Gezahagne, A. (2014). Estimating Spot Price and Smooth Forward Curve in Electricity Markets with Bayesian Penalized Spline (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25956