Robust policy evaluation from large-scale observational studies
Md Saiful Islam, Md Sarowar Morshed, Gary J Young, Md Noor-E-Alam · 2026
Read Robust policy evaluation from large-scale observational studies by Md Saiful Islam, Md Sarowar Morshed, Gary J Young, Md Noor-E-Alam (2026) 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/plos-pone-0348228.txt.
Description
Important notice first: an earlier version of this study by Md Saiful Islam and three colleagues was retracted by PLOS One. The authors had found an error in their p-value math. Editors concluded that one conclusion, drawn from a large validation dataset, was incorrect. The journal invited this corrected and republished version, which is what this entry holds. The core method was judged sound.
The work itself tackles uncertainty in matching-based policy evaluation. When several matched pairs look equally good, different algorithms can yield different verdicts. The authors propose faster algorithms that test whether a causal conclusion survives all reasonable assignments. They validate the approach on California hospital data. Applied there, the Medicare readmission penalty programme showed no causal link to rises in readmissions at other hospitals. The scalable test lets analysts check big observational studies without prohibitive computing cost.
Catalog record
| CREATOR | Md Saiful Islam, Md Sarowar Morshed, Gary J Young, Md Noor-E-Alam |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52074 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |