IISE PG&E Energy Analytics Challenge 2025: Hourly-Binned Regression Models Beat Transformers in Load Forecasting

TL;DR

The paper tackles long-horizon electricity load forecasting under data-constrained conditions using the PG&E ESD 2025 dataset. It frames the problem as 24 hourly regression tasks with PCA-based exogenous features and stacks them into full-year forecasts, comparing a broad suite of models from Piecewise/Polynomial regressions to XGBoost, RF, MLP, LSTM, and transformer-based approaches including TimeGPT. The key finding is that deep learning models, including TimeGPT, do not consistently outperform simpler methods; XGBoost with a single lag of PCA-transformed exogenous variables provides the best trade-off between accuracy and computational efficiency, while lagged features yield only marginal gains. The study underscores the importance of aligning model complexity with data availability and problem structure, suggesting practical forecasting solutions should favor interpretable, efficient models in constrained settings while keeping open the potential of hybrid DL-statistical methods for richer data regimes.

Abstract

Accurate electricity load forecasting is essential for grid stability, resource optimization, and renewable energy integration. While transformer-based deep learning models like TimeGPT have gained traction in time-series forecasting, their effectiveness in long-term electricity load prediction remains uncertain. This study evaluates forecasting models ranging from classical regression techniques to advanced deep learning architectures using data from the ESD 2025 competition. The dataset includes two years of historical electricity load data, alongside temperature and global horizontal irradiance (GHI) across five sites, with a one-day-ahead forecasting horizon. Since actual test set load values remain undisclosed, leveraging predicted values would accumulate errors, making this a long-term forecasting challenge. We employ (i) Principal Component Analysis (PCA) for dimensionality reduction and (ii) frame the task as a regression problem, using temperature and GHI as covariates to predict load for each hour, (iii) ultimately stacking 24 models to generate yearly forecasts. Our results reveal that deep learning models, including TimeGPT, fail to consistently outperform simpler statistical and machine learning approaches due to the limited availability of training data and exogenous variables. In contrast, XGBoost, with minimal feature engineering, delivers the lowest error rates across all test cases while maintaining computational efficiency. This highlights the limitations of deep learning in long-term electricity forecasting and reinforces the importance of model selection based on dataset characteristics rather than complexity. Our study provides insights into practical forecasting applications and contributes to the ongoing discussion on the trade-offs between traditional and modern forecasting methods.

Figures (9)

• Figure 1: Load and Site 5 Temperature for Year 1.
• Figure 2: Load and Site 5 Temperature for Year 2.
• Figure 3: PCA components.
• Figure 4: Weekdays vs Weekends from Load data
• Figure 5: Test Cases for XGBoost Model