CMStatistics 2021: Start Registration
View Submission - CMStatistics
B0667
Title: Clustering random intervals: A new approach based on a similarity measure Authors:  Ana Belen Ramos-Guajardo - Fundacion Universidad de Oviedo (Spain) [presenting]
Abstract: A hierarchical method for clustering random intervals is proposed. The idea is to group random intervals based on the similarity of their corresponding expected values. In this way, two random intervals will be joined if the degree of similarity of their expected values can be assumed to be greater than or equal to a certain degree. Such a similarity degree between each pair of random intervals can be analyzed by means of a two-sample similarity bootstrap test providing finally a $p$-value matrix. Thus, the higher the $p$-value obtained, the greater the similarity between both random intervals. The iterative clustering algorithm suggested comprises an objective stopping criterion that leads to statistically similar clusters that are different from each other. Lastly, a comparative simulation study and an application of the method to a real case are shown.