Comparative Analysis of ARIMA and BSTS Models for Electricity Price Forecasting in Major West Coast Cities

Authors

  • Ziyang Zhang

DOI:

https://doi.org/10.54097/beksq597

Keywords:

ARIMA model; BSTS model; electricity price forecasting.

Abstract

This study presents a detailed comparative analysis of two prominent statistical models, the AutoRegressive Integrated Moving Average (ARIMA) and Bayesian Structural Time Series (BSTS), in the context of forecasting electricity prices in major West Coast cities of the United States, namely Los Angeles, San Francisco, and Seattle. Utilizing historical electricity price data obtained from the Bureau of Labor Statistics, the research meticulously trains and tests both models, aiming to evaluate their predictive accuracy and reliability in dynamic urban energy markets. The core objective of this research is to ascertain the effectiveness of ARIMA and BSTS models in the realm of electricity price forecasting, with a particular emphasis on their applicability in urban settings characterized by rapid market evolution. Through this analysis, the study offers valuable insights that could significantly influence energy market forecasting and policy formulation. The findings are particularly relevant for stakeholders in the energy sector, including policymakers and market strategists, as they navigate the complexities and challenges of modern energy landscapes. This research contributes to a more comprehensive understanding of statistical modeling techniques in energy market analysis and provides a foundation for more informed decision-making in energy policy and strategy development.

Downloads

Download data is not yet available.

References

Cˇesnaviˇcius M. Lithuanian electricity market price forecasting model based on univariate time series analysis. Energetika (Vilnius, Lithuania), 2020, 66(1).

Elsaraiti M, Ali G, Musbah H, Merabet A, Little T. Time series analysis of electricity consumption forecasting using ARIMA model. 2021 IEEE Green Technologies Conference (GreenTech), 2021, 259-262.

Xie L. The analysis and forecasting of COVID-19 cases in the United States using Bayesian structural time series models. Biostatistics & Epidemiology (Oxford, UK), 2022, 6(1), 1-15.

Kalinina I, Bidyuk P, Gozhyj A. Construction of forecast models based on bayesian structural time series. 2022 IEEE 17th International Conference on Computer Sciences and Information Technologies (CSIT), Lviv, Ukraine, 2022, 180-184.

Carriere T, Kariniotakis G. An integrated approach for ValueOriented energy forecasting and Data-Driven Decision-Making application to renewable energy trading. IEEE Transactions on Smart Grid, 2019, 10(6), 6933-6944.

Conejo A J, Plazas M A, Espinola R, Molina A B. Dayahead electricity price forecasting using the wavelet transform and ARIMA models. IEEE Transactions on Power Systems, 2005, 20(2), 1035–1042.

Bernardi M, Petrella L. Multiple seasonal cycles forecasting model: the Italian electricity demand. Statistical Methods & Applications, 2015, 24(4), 671-695.

Rodr´ıguez Y, D´ıaz Olariaga O. Air traffic demand forecasting with a Bayesian structural time series approach. Transportation Engineering, 2023.

Cuaresma J C, Hlouskova J, Kossmeier S, et al. Forecasting electricity spot-prices using linear univariate time-series models. Applied Energy, 2004, 77(1): 87-106.

Pavlyshenko B. Bayesian regression approach for building and stacking predictive models in time series analytics. Data Stream Mining & Processing: Third International Conference, DSMP 2020, Lviv, Ukraine, August 21–25, 2020, Proceedings 3. Springer International Publishing, 2020: 486-500.

Downloads

Published

29-03-2024