Open Datseris opened 1 year ago
Found another method to construct causal graphs here:
Nice. Good catch! We should implement this too. I'm not too worried of the "quality" of the methods when it comes to a decision on incorporating them here or not. The advantage om implementing here it is that we can test it against all other methods, on any system.
Sound sgood, I agree with your argument for implementing everything.
Found another method to construct causal graphs here: https://pure.mpg.de/rest/items/item_3384208/component/file_3384211/content
But then again, the paper says that Granger Causality, that is to my knowledge one of the weakest algorithmns to use for causal graphs, performs even better than this one...
Lead author @moritz-g has the method coded here: https://github.com/moritz-g/causal_networks_physiology