Performing price scenario analysis and stress testing using quantile regression: A case study of the Californian electricity market

This paper uses quantile regression to demonstrate how electricity price distributions are linked to fundamental supply and demand variables. It investigates the California electricity market (zone SP15) for selected trading hours using data from January 8, 2013 to September 24, 2016. The approach quantifies a non-linear relationship between the fundamentals and electricity prices, just as predicted by the merit order curve. Natural gas, greenhouse gas allowance prices and load all have a positive effect on electricity prices, with the effect increasing with the quantiles. In contrast, solar production and wind production both have a negative effect on electricity prices. The effect of solar production increases with quantiles, whereas the effect of wind production decreases with quantiles. This paper also includes a stress testing case study in which a producer faces the risk of high solar and wind production, and investigates the effect on the lower tail of the price distribution. Overall, the results demonstrate how the proposed approach can be a helpful risk management tool for participants in the electricity market. (C) 2020 The Author(s). Published by Elsevier Ltd.

Automated summary

This paper uses quantile regression to demonstrate the relationship between electricity price distributions and fundamental supply and demand variables, finding that natural gas, greenhouse gas allowance prices, and load have a positive effect on electricity prices, while solar production and wind production have a negative effect. The proposed approach can serve as a helpful risk management tool for participants in the electricity market.