Conditioning on Specific Variables

In order to enforce the inclusion of specific variables in the conditioning set, add the 'add_conditionals' key in the analysis settings dictionary and provide the list of variables to include. Each variable is defined as a tuple (process index, lag), e.g. (0, 3) corresponds to the variable in process 0 with a lag of 3 samples with respect to the target.

settings = {'add_conditionals': [(0,3), (1,2)]}

By setting 'add_conditionals' to 'faes', the algorithm adds the current value of all sources to the conditioning set. This is meant to correct for instantaneous mixing in the sources (Faes, 2013, Entropy 15(1)).

settings = {'add_conditionals': 'faes'}