CMStatistics 2023: Start Registration
View Submission - CMStatistics
B1471
Title: Test for constancy of the variance in a time series Authors:  Sara Kristin Schmidt - Ruhr-University Bochum (Germany)
Max Wornowizki - TU Dortmund University (Germany)
Roland Fried - TU Dortmund University (Germany)
Herold Dehling - Ruhr-University Bochum (Germany) [presenting]
Davide Giraudo - University of Strasbourg (France)
Abstract: A novel approach is presented to test for heteroscedasticity of a non-stationary time series based on Gini's mean difference of logarithmic local sample variances. In order to analyse the large sample behaviour of the test statistic, new limit theorems are established for U-statistics of dependent triangular arrays. The asymptotic distribution of the test statistic is derived under the null hypothesis of constant variance, and it is shown that the test is consistent against a large class of alternatives, including multiple breaks. The test is applicable even in the case of non-stationary processes, assuming a locally stationary mean function.