The rEDM software package written by Hao Ye (Ye et al. in prep) is a toolbox of commonly used methods in empirical dynamics. It is based on laboratory research code developed in the early 90’s and contains methods to identify and quantify nonlinear predictability (Sugihara & May 1990, Sugihara 1994) as well as multivariate extensions thereof (Dixon et al. 1999, Ye and Sugihara 2016). In addition, rEDM implements convergent cross mapping as a test for causality (Sugihara et al. 2012), and a method for measuring changing interactions among causal variables (Deyle et al. 2016).

Access the software package here at github or here at CRAN.


A Python implementation of EDM tools providing functionality similar to the R implementation rEDM by Ye et. al. Functionality includes simplex projection (Sugihara and May 1990), sequential locally weighted global linear maps (S-map) (Sugihara 1994), multivariate embeddings (Dixon et. al. 1999), convergent cross mapping (Sugihara et. al. 2012),  and multiview embedding (Ye and Sugihara 2016).

The pyEDM package can be accessed here at github.


A State implementation of empirical dynamic modeling by Li et. al. can be accessed here.