EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

1935. Markov Chain Bootstrapping and Simulation for a Quadrivariate Stochastic Process in Energy Market Scenario Generation

Invited abstract in session TD-39: Analysis of Stochastic Models I, stream Stochastic Modelling.

Tuesday, 14:30-16:00
Room: 35 (building: 306)

Authors (first author is the speaker)

1. Cristian Pelizzari
Department of Economics and Management, University of Brescia
2. Enrico Angelelli
Metodi Quantitativi, University of Brescia
3. Paolo Falbo
Department of Economics and Management, University of Brescia
4. Alessandra Ruffini
Department of Economics and Management, University of Brescia

Abstract

Research in energy markets often requires jointly considering multiple sources of uncertainty. Among all players in the energy markets, typical sources of profit uncertainty are supply costs, weather conditions, sale prices, and demand levels. Typically, these components are not independent of one another. Moreover, the dependencies are often non-linear. In turn, this makes investment and management decisions and risk assessment complex.
In the methodological field, an approach that has proven particularly effective in modelling linear and non-linear dependencies among multiple sources of uncertainty was advanced in Cerqueti et al. (2017). The method is based on the approximation of stochastic processes through Markov chains of order k, with k belonging to the set {1,2,...}.
The present work is concerned with modelling the quadrivariate stochastic process of natural gas price, electricity demand, electricity price, and solar radiation based on the method of in Cerqueti et al. (2017). The novel aspect of the present method compared to its previous applications consists in the management of the different temporal frequencies of the four components (gas prices have a daily frequency, while the others have an hourly frequency).
The method, based on the choice of a few parameters compared to other methods, allows us to bootstrap and simulate the quadrivariate stochastic process. Statistical tests are applied to the generated scenarios to assess their goodness of fit.

Keywords

Status: accepted

Back to the list of papers