Projects per year
Abstract
Given the output of a data source taking values in a finite alphabet, we wish to estimate changepoints, that is times when the statistical properties of the source change. Motivated by ideas of match lengths in information theory, we introduce a novel nonparametric estimator which we call CRECHE (CRossings Enumeration CHange Estimator). We present simulation evidence that this estimator performs well, both for simulated sources and for real data formed by concatenating text sources. For example, we show that we can accurately estimate the point at which a source changes from a Markov chain to an IID source with the same stationary distribution. Our estimator requires no assumptions about the form of the source distribution, and avoids the need to estimate its probabilities. Further, establishing a fluid limit and using martingale arguments.
Original language  English 

Pages (fromto)  9871008 
Journal  Methodology and Computing in Applied Probability 
Volume  16 
Issue number  4 
Early online date  17 Jul 2013 
DOIs  
Publication status  Published  Dec 2014 
Fingerprint
Dive into the research topics of 'Nonparametric changepoint detection using string matching algorithms'. Together they form a unique fingerprint.Projects
 1 Finished

Efficient entropybased detection of ChangePoints in Streaming Data
1/07/10 → 1/01/11
Project: Research
Profiles

Professor Robert J Piechocki
 Department of Electrical & Electronic Engineering  Professor of Wireless Systems
 Communication Systems and Networks
Person: Academic , Member