Closed aaorsi closed 1 year ago
Hello,
There are some algorithms that can already return a weighted structure as a result of how the DAG is computed. You can take a look at LiNGAM family of algorithms. gCastle provides ICA-based LiNGAM and DirectLiNGAM, but there are more extensions, you can take a look here for an overview: https://sites.google.com/view/sshimizu06/lingam
Fantastic, thanks a lot!
Hi! Excellent library! have you considered including a measure of causal strength, like parameter learning for bayesian networks but applied to these DAGs? What would you suggest to estimate that?