Estimating a State-Space Model from Point Process Observations: A Note on Convergence
Estimating a State-Space Model from Point Process Observations: A Note on Convergence
Physiological signals such as neural spikes and heartbeats are discrete events in time, driven by continuous underlying systems. A recently introduced data-driven model to analyze such a system is a state-space model with point process observations, parameters of which and the underlying state sequence are simultaneously identified in a maximum likelihood setting using the expectation-maximization (EM) algorithm. In this note, we observe some simple convergence properties of such a setting, previously un-noticed. Simulations show that the likelihood is unimodal in the unknown parameters, and hence the EM iterations are always able to find the globally optimal solution.
1993-2001
Yuan, Ke
ae93001b-9371-4d70-88a5-ea53b2c429a9
Niranjan, Mahesan
5cbaeea8-7288-4b55-a89c-c43d212ddd4f
August 2010
Yuan, Ke
ae93001b-9371-4d70-88a5-ea53b2c429a9
Niranjan, Mahesan
5cbaeea8-7288-4b55-a89c-c43d212ddd4f
Yuan, Ke and Niranjan, Mahesan
(2010)
Estimating a State-Space Model from Point Process Observations: A Note on Convergence.
Neural Computation, 22 (8), .
Abstract
Physiological signals such as neural spikes and heartbeats are discrete events in time, driven by continuous underlying systems. A recently introduced data-driven model to analyze such a system is a state-space model with point process observations, parameters of which and the underlying state sequence are simultaneously identified in a maximum likelihood setting using the expectation-maximization (EM) algorithm. In this note, we observe some simple convergence properties of such a setting, previously un-noticed. Simulations show that the likelihood is unimodal in the unknown parameters, and hence the EM iterations are always able to find the globally optimal solution.
Text
yn_nc_2010.pdf
- Version of Record
Restricted to Registered users only
Request a copy
More information
Published date: August 2010
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 271146
URI: http://eprints.soton.ac.uk/id/eprint/271146
PURE UUID: 801902fc-8961-4f0b-9f79-71965627ce92
Catalogue record
Date deposited: 21 May 2010 13:29
Last modified: 15 Mar 2024 03:29
Export record
Contributors
Author:
Ke Yuan
Author:
Mahesan Niranjan
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics