Reassessing Hierarchical Forecasting: Supporting Data and Code

Full reproducibility package for the article “Reassessing Hierarchical Forecasting: An Empirical Null Result” submitted to the International Journal of Forecasting. The archive contains three top-level folders: DATA/ (prepared WalMart M5 series, pooled and per-metric tables, and comparison summaries), R/ (all R scripts used for data preparation, forecasting, and result aggregation), and SQL/ (PostgreSQL table definitions and analysis queries). Together, these files recreate the master result tables T01–T15 and all summary statistics reported in the manuscript.