Scikit Convergent Cross Mapping (skccm) can be used as a way to detect causality between time series.
For a quick explanation of this package, I suggest checking out the Documentation as well as the wikipedia article on convergent cross mapping . Additionally, Dr. Sugihara's lab has produced some good summary videos about the topic:
- Time Series and Dynamic Manifolds
- Reconstructed Shadow Manifold
- State Space Reconstruction: Convergent Cross Mapping
For a more complete background, I suggest checking out the following papers:
- Detecting Causality in Complex Ecosystems by Sugihara
- Distinguishing time-delayed causal interactions using convergent cross mapping by Ye
Sugihara also has a good talk about about Correlation and Causation