A [selected,pos="-2.200,-1.520"]
B [pos="1.400,-1.460"]
D [outcome,pos="1.400,1.621"]
E [exposure,pos="-2.200,1.597"]
Z [adjusted,pos="-0.300,-0.082"]
A -> E
A -> Z [pos="-0.791,-1.045"]
B -> D
B -> Z [pos="0.680,-0.496"]
E -> D
Your canvas is empty
Click anywhere to add a variable.
Then click two variables in sequence to draw an arrow.
New here? Try Help → Build a DAG step-by-step, or an Example from the menu above.
Upload a data file to automatically suggest a causal graph using the PC algorithm (skeleton + collider orientation) combined with LiNGAM-inspired pairwise orientation (non-Gaussianity of residuals).
Python pickle (.pkl): export first with df.to_csv('data.csv', index=False)
Discovered edges
A clean, manuscript-ready redraw of your diagram following the usual DAG conventions — conditioned variables in boxes, unobserved variables dashed. Export as scalable SVG (best for journals & editing in Illustrator/Inkscape), high-resolution PNG, or PDF.