Detection with the scan and the average likelihood ratio
From MaRDI portal
Publication:4908798
Abstract: We investigate the performance of the scan (maximum likelihood ratio statistic) and of the average likelihood ratio statistic in the problem of detecting a deterministic signal with unknown spatial extent in the prototypical univariate sampled data model with white Gaussian noise. Our results show that the scan statistic, a popular tool for detection problems, is optimal only for the detection of signals with the smallest spatial extent. For signals with larger spatial extent the scan is suboptimal, and the power loss can be considerable. In contrast, the average likelihood ratio statistic is optimal for the detection of signals on all scales except the smallest ones, where its performance is only slightly suboptimal. We give rigorous mathematical statements of these results as well as heuristic explanations which suggest that the essence of these findings applies to detection problems quite generally, such as the detection of clusters in models involving densities or intensities or the detection of multivariate signals. We present a modification of the average likelihood ratio that yields optimal detection of signals with arbitrary spatial extent and which has the additional benefit of allowing for a fast computation of the statistic. In contrast, optimal detection with the scan seems to require the use of scale-dependent critical values.
Recommendations
Cited in
(32)- Multiscale scanning in inverse problems
- Generalized multiple change-point detection in the structure of multivariate, possibly high-dimensional, data sequences
- Optimal change-point detection and localization
- Detecting a target in very noisy data from mutliple looks
- Minimax and adaptive tests for detecting abrupt and possibly transitory changes in a Poisson process
- Detecting multiple generalized change-points by isolating single ones
- Cluster detection in networks using percolation
- Two-stage data segmentation permitting multiscale change points, heavy tails and dependence
- A change-point problem and inference for segment signals
- Multiscale change-point segmentation: beyond step functions
- Cross-validation for change-point regression: pitfalls and solutions
- Bump detection in the presence of dependency: does it ease or does it load?
- Wild binary segmentation for multiple change-point detection
- Narrowest-Over-Threshold Detection of Multiple Change Points and Change-Point-Like Features
- Optimal detection of multi-sample aligned sparse signals
- Variational multiscale nonparametric regression: smooth functions
- Localising change points in piecewise polynomials of general degrees
- FDR-control in multiscale change-point segmentation
- Identifying the support of rectangular signals in Gaussian noise
- A comparison of single and multiple changepoint techniques for time series data
- Distribution-free detection of structured anomalies: permutation and rank-based scans
- Frequency domain bootstrap methods for random fields
- Optimal inference with a multidimensional multiscale statistic
- Exact asymptotics for the scan statistic and fast alternatives
- Testing and estimation for clustered signals
- Multiscale change point inference. With discussion and authors' reply
- Optimal detection of a jump in the intensity of a Poisson process or in a density with likelihood ratio statistics
- Persistence barcodes versus Kolmogorov signatures: detecting modes of one-dimensional signals
- Narrowest Significance Pursuit: Inference for Multiple Change-Points in Linear Models
- Structured correlation detection with application to colocalization analysis in dual-channel fluorescence microscopic imaging
- Large-scale inference with block structure
- Univariate mean change point detection: penalization, CUSUM and optimality
This page was built for publication: Detection with the scan and the average likelihood ratio
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4908798)
