|
Abstract
|
Multiple change-point problems can be found in many areas of science and engineering. To detect all change-points in a data sequence is of great importance. A statistical analysis without considering their existence may lead to an incorrect or improper conclusion. We will present some numerical examples to illustrate a change-point problem, and show the importance to include a change-point in data modeling. Owing to the rapid development in model selection methods, a multiple change-point detection method can be built upon by converting a multiple change-point detection problem into a variable selection problem via proper segmentation of a data sequence. We will discuss recent developments in multiple change-point detection using this methodology and their applications in real problems.
|
