An equilibrium approach to estimate electricity forward prices in an illiquid market
Authors: Felipe Van de Sande Araujo, Cristina Spineti Luz, Leonardo Lima Gomes, Luiz Eduardo Teixeira Brandão The files provided here are the calculations for the paper with the same title. They are organized in the following way: Data - contain datasets of forecasted spot prices for electricity in SE/CO region of Brazil for the year 2020, 2021 and 2022, as well as estimated forward price for electricity for the same years. This data is loaded by the R scripts. Plots - contain the plots generated by these scripts. The main level contain the scripts that can be executed, provided the data is organized in the same structure as this repository. All code was developed for the article. Paper Abstract: The existence of electricity forward prices provides valuable price discovery, informing operational and investment decisions (that otherwise need to rely on other authoritative sources, such as estimation by market experts). However, electricity forward markets are plagued by low liquidity and by an absence of publicly available transaction prices. We propose a novel equilibrium approach to estimating forward prices. A contributing aspect here is the use of new preference functions based on mixed conditional Value-at-Risk; we are able to characterize the level of risk aversion on the buyer and seller side. The framework is tested with sets of spot prices within the context of the Brazilian market, where most forward transactions are conducted over the counter. Therefore, price information is obtained through research firms. The forward prices estimated with our approach are aligned with those specialists, and we conclude that this framework is a useful analytical tool even in an illiquid market. Furthermore, our results show the existence of a positive relationship between the skewness of the distribution of the spot prices and the estimated forward prices, and also a negative relationship between the variance of the spot prices and the estimated forward prices which verify that Bessembinder and Lemmon’s (2002) first two hypotheses hold in the Brazilian context. Keywords: electricity forward prices, bilateral contracts, multi-agent equilibrium, mixed CVaR, Optimization.
Steps to reproduce
The code was last run in R 4.1.1 All R files must be run in the working directory and the folder structure must be preserved. Libraries required for chart generation: plotly (for interactive and non-interactive plots) and viridis (optimized for colorblind visualization support).