AN INFORMATION MEASURE OF f-DIVERGENCE FAMILY AND ITS PROPERTIES
By
K.C.Jain1 and Devendra Singh2
1Department of Mathematics, IIIT Kota, JLN Marg, MNIT Campus, Jaipur-302017, Rajasthan, India
2Department of Mathematics, Malaviya National Institute of Technology, Jaipur-302017, Rajasthan, India
Email: devendra.mnit2013@gmail.com , 2013rma9061@mnit.ac.in
(Received : November 24, 2017 ; Revised: September 03, 2018)
Abstract
A non-parametric symmetric information divergence measure is proposed. This information measure belongs to the family of Csiszrs f-divergences. Its properties are studied and discussed. Further, we have derived some new bounds for this information divergence measure in terms of some recognized divergence measures based on two distinct probability distributions.
2010 Mathematics Subject Classification: 94A17, 26D15.
Keywords and phrases: Non-parametric measure, Kullback-Leibler Measure, Csiszrs f-divergence, Information measure, Triangular Discremination, Symmetric chi-square divergence.