High-dimensional change-point detection with sparse alternatives
By Farida N. Enikeeva, Zaïd Harchaoui. 2013
63 citations recorded by the source corpus.
We consider the problem of detecting a change in mean in a sequence of Gaussian vectors. Under the alternative hypothesis, the change occurs only in some subset of the components of the vector. We propose a test of the presence of a change-point that is adaptive to the number of changing components. Under the assumption that the vector dimension tends to infinity and the length of the sequence grows slower than the dimension of the signal, we obtain the detection boundary for this problem and prove its rate-optimality.
Subfields: Statistics and Probability
View Paper: https://www.semanticscholar.org/paper/2329babee8777224c86362a149f3f8d1f054e726
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Statistics and Probability
Citations in the corpus
- Selective review of offline change point detection methods (2020, methodology): “In [149], a procedure for single 955 mean-shifts is described: the estimation is shown to be consistently for T and d 956 increasing to infinity, and under certain conditions (in particular, the number of 957 changing components must be sufficiently large) Also, their method has a linear 958 complex”
- Finite sample change point inference and identification for high‐dimensional mean vectors (2017, methodology): “However, in such case, the bootstrap CUSUM test does not achieve the detection boundary and the minimax separation rate (Enikeeva & Harchaoui, 2019).”
- Change-point detection in panel data via double CUSUM statistic (2016, methodology): “For the purpose of testing only, it is reasonable to combine the two tests based on D0 m and D 1/2 m as in Enikeeva and Harchaoui (2015), namely, (T 0 1,T > π0 n,T ) ∨ (T 1/2 1,T > π 1/2 n,T ).”