Tools for forecast reconciliation: FoReco 0.2
FoReco is an R package designed for point forecast reconciliation of a multiple linearly constrained (e.g. hierarchical/grouped) time series. The present release (0.1.1, https://CRAN.R-project.org/package=FoReco) deals with cross-sectional (Hyndman et al., 2011), temporal (Athanasopoulos et al., 2017), and cross-temporal (Kourentzes and Athanasopoulos, 2019, Di Fonzo and Girolimetto, 2020) forecast reconciliation procedures. Projection and structural approaches have been considered to better exploit the linear relationships linking the data. Almost all of the state-of-the-art reconciliation procedures were considered, and a powerful tool to guarantee non-negativity of the reconciled forecasts was made available.
In the new FoReco 0.2.0, some significant updates have been considered: first, building upon and extending a recent proposal by Hollyman et al. (2021), a new forecast combination based forecast reconciliation procedure is considered, with either exogenous or endogenous intermediate level constraints. Level Conditional Coherent (LCC) forecast reconciliation for elementary hierarchies, and the Combined Conditional Coherent forecast reconciliation approach are now available in the new command lccrec(). In addition, the new release, (i) besides the non-negative option, now permits to impose linear inequalities bounds on all the reconciled forecasts, which may be very useful in many practical situations, (ii) offers a wider freedom in choosing the covariance matrices (also different along the forecast horizon), and (iii) let the user define a customized subset of the temporal aggregation orders to be used in the reconciliation.
References
Athanasopoulos, G., Hyndman, R.J., Kourentzes, N., Petropoulos, F. (2017), Forecasting with Temporal Hierarchies, European Journal of Operational Research, 262, 1, 60-74.
Di Fonzo, T., Girolimetto, D. (2020), Cross-Temporal Forecast Reconciliation: Optimal Combination Method and Heuristic Alternatives, https://arxiv.org/abs/2006.08570.
Hollyman, R., Petropoulos, F., Tipping, M.E. (2021), Understanding Forecast Reconciliation, European Journal of Operational Research (in press).
Kourentzes, N., Athanasopoulos, G. (2019), Cross-temporal coherent forecasts for Australian tourism, Annals of Tourism Research, 75, 393-409.
Hyndman, R.J., Ahmed, R.A., Athanasopoulos, G., Shang, H.L. (2011), Optimal combination forecasts for hierarchical time series, Computational Statistics & Data Analysis, 55, 9, 2579-2589.