A review of instrumental variable estimators for Mendelian randomization.
Burgess S, Small DS, Thompson SG. · 2017
Read A review of instrumental variable estimators for Mendelian randomization. by Burgess S, Small DS, Thompson SG. (2017) 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/europmc-pmc5642006.txt.
Description
Burgess, Small, and Thompson present a methods review in Statistical Methods in Medical Research. It compares estimators used in Mendelian randomization. That approach uses genetic variants as instrumental variables. A valid instrument links to the exposure but not to confounders. It also must reach the outcome only through the exposure. Under those rules, causal effects can be drawn from observational data.
Four families of methods get compared. They are ratio, two-stage, likelihood-based, and semi-parametric methods. The authors weigh bias, coverage, and efficiency for each. Weak instruments get special attention, since they distort estimates. A score-based approach helps when many weak instruments are used. Practical advice guides readers on choosing a method.
Catalog record
| CREATOR | Burgess S, Small DS, Thompson SG. |
|---|---|
| YEAR | 2017 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 95 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |