This talk has captions. You can remove these by pressing CC on the video toolbar.  

Name: Jorge Castillo-Mateo

Talk Title: Nonparametric Changepoint Detection Tests Based on the Breaking of Records

Abstract: Three new nonparametric statistics are introduced for the changepoint detection problem. They are based on functions of the numbers of upper or/and lower records which occur in a series. The asymptotic Kolmogorov distribution of the test statistics is obtained from a characterization based on Wiener processes and Brownian bridges. A Monte Carlo study of size, power and changepoint estimate is developed. The methods are illustrated by the analysis of climate change through the record occurrence of temperature time series. Record statistics are not a common approach to extreme value analysis, but this work shows some of their potential interests.

This talk is a contributed talk at EVA 2021. View the programme here.