A Sparse Linear Model and Significance Test for Individual Consumption Prediction

Accurate prediction of user consumption is a key part not also in understanding consumer flexibility and behavior patterns, but in the design of robust and efficient energy saving programs as well. Existing prediction methods usually have high relative errors that can be larger than 40\%. In this paper, we propose a method to improve prediction accuracy of individual users by exploring sparsity in historical data and leveraging predictive relationship between different users. Sparsity is captured by popular least absolute shrinkage and selection (LASSO) estimator, while user selection is formulated as an optimal hypothesis testing problem and solved via a covariance test. We provide extensive simulation against well-known techniques such as support vector machine (SVM), principle component analysis (PCA) and random forest (RF) using real world data. Simulation results demonstrate that our proposed methods are operationally efficient, interpretable, and achieves optimal prediction performance.