Weighted Median
Resources
Code (Language: R) Website Paper (Pub Date: 2016-04-07)
Other URLS: https://cran.r-project.org/web/packages/MendelianRandomization/index.html
Method Contact: Jack Bowden ( J.Bowden2@exeter.ac.uk)
Entry Contact: Gibran Hemani ( g.hemani@bristol.ac.uk)
Context
Analysis Type: UVMR MVMR Network MR Bi-Directional MR Non-Linear
Input Data Types: Ind/Ind Ind/SS SS/Ind SS/SSfamily
Exposure Trait Types: Quantitative Binary Time to event
Outcome Trait Types: Quantitative Binary Time to event
Assumptions and Sources of Bias
| Source of Bias | Addressed |
|---|---|
| Weak instruments | ✗ |
| Winner’s curse | ✗ |
| Sample overlap | ✗ |
| Uncorrelated horizontal pleiotropy | ✓ |
| Correlated horizontal pleiotropy | ✗ |
| Ancestry differences in samples | ✗ |
| Residual confounding in GWAS | ✗ |
| Cross-trait assortative mating | ✗ |
| Index-event/conditioning on heritable trait | ✗ |
Description
A novel weighted median estimator for combining data on multiple genetic variants into a single causal estimate. This estimator is consistent even when up to 50% of the information comes from invalid instrumental variables. In a simulation analysis, it is shown to have better finite-sample Type 1 error rates than the inverse-variance weighted method, and is complementary to the recently proposed MR-Egger (Mendelian randomization-Egger) regression method.