On the Bias of Directed Information Estimators

arXiv1902.00580MaRDI QIDQ6313456

Author name not available (Why is that?)

Publication date: 1 February 2019

Abstract: When estimating the directed information between two jointly stationary Markov processes, it is typically assumed that the recipient of the directed information is itself Markov of the same order as the joint process. While this assumption is often made explicit in the presentation of such estimators, a characterization of when we can expect the assumption to hold is lacking. Using the concept of d-separation from Bayesian networks, we present sufficient conditions for which this assumption holds. We further show that the set of parameters for which the condition is not also necessary has Lebesgue measure zero. Given the strictness of these conditions, we introduce a notion of partial directed information, which can be used to bound the bias of directed information estimates when the directed information recipient is not itself Markov. Lastly we estimate this bound on simulations in a variety of settings to assess the extent to which the bias should be cause for concern.

Has companion code repository: https://github.com/gabeschamberg/directed_info_bias

This page was built for publication: On the Bias of Directed Information Estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6313456)