A pdf-Free Change Detection Test Based on Density Difference Estimation
Authors:
- 16 November 2016
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 29 (2) , 324-334
Abstract
The ability to detect online changes in stationarity or time variance in a data stream is a hot research topic with striking implications. In this paper, we propose a novel probability density function-free change detection test, which is based on the least squares density-difference estimation method and operates online on multidimensional inputs. The test does not require any assumption about the underlying data distribution, and is able to operate immediately after having been configured by adopting a reservoir sampling mechanism. Thresholds requested to detect a change are automatically derived once a false positive rate is set by the application designer. Comprehensive experiments validate the effectiveness in detection of the proposed method both in terms of detection promptness and accuracy.Keywords
Funding Information
- National Natural Science Foundation of China (61273136, 61573353, 61533017, 61603382)
This publication has 41 references indexed in Scilit:
- Density-Difference EstimationNeural Computation, 2013
- Change-point detection in time-series data by relative density-ratio estimationNeural Networks, 2013
- Exponentially weighted moving average charts for detecting concept driftPattern Recognition Letters, 2012
- Nonparametric Monitoring of Data Streams for Changes in Location and ScaleTechnometrics, 2011
- Least-squares two-sample testNeural Networks, 2011
- A Multivariate Change-Point Model for Statistical Process ControlTechnometrics, 2006
- A Multivariate Two-Sample Test Based on the Number of Nearest Neighbor Type CoincidencesThe Annals of Statistics, 1988
- Multivariate Two-Sample Tests Based on Nearest NeighborsJournal of the American Statistical Association, 1986
- Random sampling with a reservoirACM Transactions on Mathematical Software, 1985
- The Kolmogorov-Smirnov Test for Goodness of FitJournal of the American Statistical Association, 1951