Preliminary Bias Analysis of Experiments for Multi-Item Multi-Period Inventory Control Policies
Xinqi Chen, Xingyu Bai, Zeyu Zheng, Nian Si · 2025
Read Preliminary Bias Analysis of Experiments for Multi-Item Multi-Period Inventory Control Policies by Xinqi Chen, Xingyu Bai, Zeyu Zheng, Nian Si (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-11996.txt.
Description
Xinqi Chen, Xingyu Bai, Zeyu Zheng, and Nian Si study why A/B tests mislead in inventory systems, in a paper from twenty twenty-five. Randomized experiments are the gold standard for judging a policy, yet stock managers rarely use them. The team analyzed two standard designs in systems holding many items over many periods, with lost sales and capacity limits.
Both designs suffer systematic bias from interference: switchback tests leak effects across time, while item-level tests let items cannibalize each other under tight capacity. The authors derive which way the bias points under mild conditions. They then propose a paired design across items and time, and check it in simulations and on real fresh-retail data.
Catalog record
| CREATOR | Xinqi Chen, Xingyu Bai, Zeyu Zheng, Nian Si |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-11996