Using Generative Pre-trained Transformers (GPT) for Electricity Price Trend Detection in the Spanish Market
Authors: Alberto Menéndez Medina, José Antonio Heredia Álvaro
Affiliation: Cátedra Industria 4.0 Universitat Jaume I, Castellón, Spain
The electricity market in Spain plays a significant role in the nation's economy and sustainability efforts. Accurate energy price prediction is crucial, influencing climate goals, energy security, and economic stakeholders. This research analyzes the impact of news and expert reports using GPT tools, like OpenAI's ChatGPT, on electricity price trend prediction for the Spanish market.
Two training and modeling approaches of Generative Pre-trained Transformers (GPT) with specialized news feeds specific to the Spanish market are proposed: In-context example prompts and fine-tuned GPT models. Integrating GPT insights into electricity price trend forecasting can result in more precise predictions and a deeper understanding of market dynamics.
The forecast of the medium to long-term price trend of the electricity market is subject to considerable uncertainty due to the influence of multiple complex factors. Analyzing the evolution of the Spanish electricity market from 2018 to 2023 reveals key factors behind heightened volatility. The COVID-19 pandemic and the current energy crisis have added complexity to the energy landscape, impacting energy markets globally.
Two primary paradigms emerge regarding integrating private data and tailoring these models to specific tasks: context learning and fine-tuning. These paradigms unlock the potential to enhance model performance, adapt to domain-specific nuances, and leverage the unique characteristics of proprietary datasets.
Paradigm 1: In-Context Learning
In-context learning allows users to interact with GPT models by providing context-specific instructions or queries. This approach enables the customization of model responses without directly modifying the model.
Paradigm 2: Fine-Tuning
Fine-tuning represents a deeper level of customization, where the pre-trained GPT model is adapted to perform specific tasks or excel domains. The model is further trained on domain-specific data, including proprietary datasets, specialized text, or even structured information.
From the results of analyzing the news and articles, insights provided by this analysis are grouped into short-term and mid/long-term, calculating the average impact for each interval and the sign indicating Direction. Future research directions include enhanced contextual understanding, incorporation of GPT-calculated features into multivariate time series prediction models, and automatic generation of reports analyzing the recent evolution of electricity market price.
This research explored the crucial realm of energy price prediction within the Spanish electricity market, utilizing GPT models to create a new approach to energy price trend forecasts. Continued exploration and optimization of OpenAI's capabilities are essential to unlock their full potential in energy price forecasting.