pankajcivil / dependency-redundancy

Code for the example of our WRR commentary

dependency-redundancy

Code for the example of the WRR paper:

Dependency and redundancy: how information theory untangles three variable interactions in environmental data

By Steven V, Weijs, Akhil Kumar, and Hossein Foroozand

Submitted to Water Resources Research, 2018

This file is provided to further explore the example, which is a difficult test case for partial information decomposition approaches. The idea of this example is that it provides a case where there is dependence between sources, but only over a range where the sources do not jointly inform the target variable T. Conversely, over the range where T is a sum of both sources, the sources are independent. This is to highlight that dependency of variables is not a guarantee for redundant influencing of a target, and to aid in developing methods of estimating redundancy that can handle this behaviour. For more information, please contact the corresponding author of the article mentioned above.

To install: add functions to the path that were downloaded from the "TIPNet" GIT repository This can be downloaded from the repository: https://github.com/HydroComplexity/TIPNet the master folder should be added as a subfolder to the path of this script the functions used are in v2, which is here added to the path: TIPNet-master\TIPNet_v2\Functions\

Note1: , this code is relating to the paper: Goodwell, A. E., & Kumar, P. (2017a). Temporal information partitioning: Characterizing synergy, uniqueness, and redundancy in interacting environmental variables. Water Resources Research, 53(7), 5920–5942. doi:10.1002/2016WR020216 Note2: the only functions used from the TIPNets toolbox by Goodwell and Kumar are: compute_pdf_fixedbins.m and compute_info_measures.m.

These can be found under the follwing direct links:

https://github.com/HydroComplexity/TIPNet/blob/master/TIPNet_v2/Functions/compute_pdf_fixedbins.m

https://github.com/HydroComplexity/TIPNet/blob/master/TIPNet_v2/Functions/compute_info_measures.m