Fisher information for a partially observable simple birth process

Title: Fisher information for a partially observable simple birth process
Creator: Bean, Nigel G.; Eshragh, Ali; Ross, Joshua V.
Relation: ARC.DP110101929
Relation: Communications in Statistics: Theory and Methods Vol. 45, Issue 24, p. 7161-7183
Publisher Link: http://dx.doi.org/10.1080/03610926.2014.978024
Publisher: Taylor & Francis
Resource Type: journal article
Date: 2016
Description: In this paper, we study the Fisher Information for the birth rate of a partially observable simple birth process involving n observations. We suppose that at each observation time, each individual in the population can be observed independently with known fixed probability p. Finding an analytical form of the Fisher Information in general appears intractable. Nonetheless, we find a very good approximation for the Fisher Information by exploiting the probabilistic properties of the underlying stochastic process. Both numerical and theoretical results strongly support the latter approximation and confirm its high level of accuracy.
Subject: continuous-time Markov chains; Fisher information; simple birth process
Identifier: http://hdl.handle.net/1959.13/1319389
Identifier: uon:23848
Identifier: ISSN:0361-0926
Language: eng
Reviewed

		Thumbnail	File	Description	Size	Format