On the ESD 2025 PG&E dataset, hourly-binned XGBoost models with PCA weather covariates achieved lower MAPE than transformer, LSTM, TFT, and TimeGPT models in day-ahead annual load forecasting.
Staff report on data needs for electricity system planning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
IISE PG&E Energy Analytics Challenge 2025: Hourly-Binned Regression Models Beat Transformers in Load Forecasting
On the ESD 2025 PG&E dataset, hourly-binned XGBoost models with PCA weather covariates achieved lower MAPE than transformer, LSTM, TFT, and TimeGPT models in day-ahead annual load forecasting.