M-Nauta / TCDF

Temporal Causal Discovery Framework (PyTorch): discovering causal relationships between time series
GNU General Public License v3.0
482 stars 106 forks source link

Assumptions #6

Closed MarkusEvo closed 3 years ago

MarkusEvo commented 3 years ago

I want to use your library to check for causal relationships between stocks (as a feature selection process for grammatical evolution because I want to evolve trading Systems). My question is: What are the underlying assumptions that you are using? Like causal sufficiency etc. Did I understand it correctly that since you make it possible to detect hidden confounders I can ignore causal sufficiency?