Additive stacking for disaggregate electricity demand forecasting

DOI10.1214/20-AOAS1417zbMATH Open1478.62373arXiv2005.10092OpenAlexW3180869686MaRDI QIDQ2245149FDOQ2245149

Yannig Goude, Christian Capezza, Simon N. Wood, Biagio Palumbo, Matteo Fasiolo

Publication date: 15 November 2021

Published in: The Annals of Applied Statistics (Search for Journal in Brave)

Abstract: Future grid management systems will coordinate distributed production and storage resources to manage, in a cost effective fashion, the increased load and variability brought by the electrification of transportation and by a higher share of weather dependent production. Electricity demand forecasts at a low level of aggregation will be key inputs for such systems. We focus on forecasting demand at the individual household level, which is more challenging than forecasting aggregate demand, due to the lower signal-to-noise ratio and to the heterogeneity of consumption patterns across households. We propose a new ensemble method for probabilistic forecasting, which borrows strength across the households while accommodating their individual idiosyncrasies. In particular, we develop a set of models or 'experts' which capture different demand dynamics and we fit each of them to the data from each household. Then we construct an aggregation of experts where the ensemble weights are estimated on the whole data set, the main innovation being that we let the weights vary with the covariates by adopting an additive model structure. In particular, the proposed aggregation method is an extension of regression stacking (Breiman, 1996) where the mixture weights are modelled using linear combinations of parametric, smooth or random effects. The methods for building and fitting additive stacking models are implemented by the gamFactory R package, available at https://github.com/mfasiolo/gamFactory.

Full work available at URL: https://arxiv.org/abs/2005.10092

Recommendations

• Hierarchical probabilistic forecasting of electricity demand with smart meter data
• Disaggregating time series on multiple criteria for robust forecasting: the case of long-term electricity demand in Greece
• Using combined forecasts with changing weights for electricity demand profiling
• Forecasting electricity consumption by aggregating specialized experts
• Triple seasonal methods for short-term electricity demand forecasting

zbMATH Keywords

ensemble methodsgeneralised additive modelsprobabilistic forecastelectricity demand forecastingregression stacking

Mathematics Subject Classification ID

Applications of statistics in engineering and industry; control charts (62P30) Inference from stochastic processes and prediction (62M20)

Cites Work

• Greedy function approximation: A gradient boosting machine.
• Title not available (Why is that?)
• Regression Quantiles
• Forecasting electricity consumption by aggregating specialized experts
• Combining predictive distributions
• Title not available (Why is that?)
• Prediction, Learning, and Games
• Dynamic Bayesian predictive synthesis in time series forecasting
• Generalized Additive Models for Location, Scale and Shape
• Title not available (Why is that?)
• Stacked regressions
• Combining Estiamates in Regression and Classification
• Combining Probability Forecasts
• Using stacking to average Bayesian predictive distributions (with discussion)
• Bayesian Nonparametric Calibration and Combination of Predictive Distributions
• Scalable Visualization Methods for Modern Generalized Additive Models
• Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Models

Cited In (3)

• Title not available (Why is that?)
• Optimized clusters for disaggregated electricity load forecasting
• gamFactory

Uses Software

• R
• GAMLSS
• GitHub
• mgcViz

This page was built for publication: Additive stacking for disaggregate electricity demand forecasting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2245149)