Fig. 3From: Overlapping-sample Mendelian randomisation with multiple exposures: a Bayesian approachFlowchart of imputations and merge of datasets A,B and C into a single complete dataset. Solid rectangles in Steps (a) and (b) denote observed data. Dashed rectangles with Y∗(t) and X∗(t) in Step (b) represent imputed values of Y and X respectively in the specific tth iteration. By merging all the observed and imputed data in A,B,C, we obtain a single complete dataset in Step (c)Back to article page