95% confidence interval for causal estimate in MR-PRESSO

[Tahzeeb]

Hi

I was wondering if there is a way to calculate 95% CI for the causal estimates in 'mr_presso' along with or right after running this function, like we do for 'lm' etc?

Many thanks and kind regards
Tahzeeb

[marieverbanck]

Hi @Tahzeeb,
it could be implemented indeed. You can do it manually using the causal estimate and its standard error and quantiles of the corresponding Student's t-distribution (with degrees of freedom equal to number of IVs - number of exposures).
Best,
Marie