Grouped time series

[edxu96]

The idea is to group products and stores into similar product and regions, for which aggregate forecasts are generated and used to determine overall seasonality and trend, which are then spread down reconciled using a Top-Down approach with the baseline forecasts generated for each individual sku/store combination.

Our method is based on independently forecasting all series at all levels of the hierarchy and then using a regression model to optimally combine and reconcile these forecasts. [hyndman2011optimal]

• 10 Forecasting hierarchical or grouped time series, fpp3
• Product Demand Forecasting for Thousands of Products Across Multiple Stores, crossvalidated
• It seems that the primary interest is in aggregated forecasts.

---

• 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.

[edxu96]

Bottom-Up Approach

An advantage of this approach is that we are forecasting at the bottom-level of a structure, and therefore no information is lost due to aggregation. On the other hand, bottom-level data can be quite noisy and more challenging to model and forecast. [FPP3]

[edxu96]

Top-Down Approach

Top-down approaches involve first generating forecasts for the Total series, and then disaggregating these down the hierarchy. This only works with strictly hierarchical data, not with grouped or mixed aggregation structures. [FPP3]