On some change-point tests based on the Laplace transform

Abstract

This talk comprises two parts. In the first part, we focus on the univariate case. Following the construction in [1], we present a novel class of tests for change-point detection based on the Laplace transform. These novel tests have favorable properties compared to the competitors in [1]. The asymptotic properties of the novel tests are discussed, and they are applied to two datasets from the fields of meteorology and macroeconomics. The second part of the study is dedicated to change-point inference for symmetric positive definite random matrices. In [3], the change-point inference for random matrices orthogonally invariant in distribution was outlined. We modify the test in [2] to test for a change-point in distribution. We present the finite sample properties of the novel test. The talk concludes with some possible directions for further research.