.png){kind=icon}
.png){kind=icon}
Return
Nonparametric two-sample test for random objects based on data depth
.png){kind=icon}
PRE
AI
DOI：10.1080/02331888.2025.2578642.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The rise of non-Euclidean data necessitates more versatile statistical tools. While two-sample testing is common, most existing methods are confined to specific data types or disparities. To address this gap, we propose an innovative, data depth-based two-sample test for random objects in metric spaces. We rigorously evaluate the test via simulation studies, examining its power and consistency against location and scale differences. Results demonstrate that our method exhibits strong finite-sample performance for vectors and probability density functions, often proving superior or comparable to existing techniques in effectiveness.
Keywords:
Data depth
density function data
two-sample test
Journal
S
IF:
1
Papers:
81
Citations:
0