Bayesian Intermittent Demand Forecasting for Large Inventories
Abstract
We present a scalable and robust Bayesian method for demand forecasting in the context of a large e-commerce platform, paying special attention to intermittent and bursty target statistics. Inference is approximated by the Newton-Raphson algorithm, reduced to linear-time Kalman smoothing, which allows us to operate on several orders of magnitude larger problems than previous related work. In a study on large real-world sales datasets, our method outperforms competing approaches on fast and medium moving items.
Cite
Text
Seeger et al. "Bayesian Intermittent Demand Forecasting for Large Inventories." Neural Information Processing Systems, 2016.Markdown
[Seeger et al. "Bayesian Intermittent Demand Forecasting for Large Inventories." Neural Information Processing Systems, 2016.](https://mlanthology.org/neurips/2016/seeger2016neurips-bayesian/)BibTeX
@inproceedings{seeger2016neurips-bayesian,
title = {{Bayesian Intermittent Demand Forecasting for Large Inventories}},
author = {Seeger, Matthias W and Salinas, David and Flunkert, Valentin},
booktitle = {Neural Information Processing Systems},
year = {2016},
pages = {4646-4654},
url = {https://mlanthology.org/neurips/2016/seeger2016neurips-bayesian/}
}