A Self-supervised Approach to Hierarchical Forecasting with Applications to Groupwise Synthetic Controls
This work addresses forecasting challenges in hierarchical data for applications like synthetic controls, offering a more integrated approach compared to existing post-hoc methods.
The paper tackles the problem of forecasting time series with hierarchical structure by proposing a new loss function that integrates reconciliation into the maximum likelihood objective, resulting in improved accuracy and proper confidence intervals. The method shows significant improvement over the state-of-the-art in synthetic data evaluations.
When forecasting time series with a hierarchical structure, the existing state of the art is to forecast each time series independently, and, in a post-treatment step, to reconcile the time series in a way that respects the hierarchy (Hyndman et al., 2011; Wickramasuriya et al., 2018). We propose a new loss function that can be incorporated into any maximum likelihood objective with hierarchical data, resulting in reconciled estimates with confidence intervals that correctly account for additional uncertainty due to imperfect reconciliation. We evaluate our method using a non-linear model and synthetic data on a counterfactual forecasting problem, where we have access to the ground truth and contemporaneous covariates, and show that we largely improve over the existing state-of-the-art method.