MR Dictionary

Maximum likelihood-based MR

Similar to the inverse variance weighted (IVW) method with fixed effects) but uses a maximum likelihood method, where the genetic effects on the exposure and outcome are modelled directly as a bivariate normal distribution. 

The method requires a term which represents the proportion of sample overlap between the exposure and outcome datasets. It also requires an iterative estimating procedure that can occasionally fail to converge. This method is largely equivalent to IVW and is not commonly used. 

References

Other terms in 'Two-sample MR methods':