Testing for multimodality
Exploratory methods for investigating the structure of a population, such as density estimation, often suggest the presence of modes. These modes can either be spurious artifacts of sampling fluctuation or else be actual features of the underlying distribution. This paper considers hypothesis testin...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
1994
|
| In: |
Computational statistics & data analysis
Year: 1994, Volume: 18, Issue: 5, Pages: 499-512 |
| DOI: | 10.1016/0167-9473(94)90080-9 |
| Online Access: | Verlag, Volltext: http://dx.doi.org/10.1016/0167-9473(94)90080-9 Verlag, Volltext: http://www.sciencedirect.com/science/article/pii/0167947394900809 
|
| Author Notes: | N.I. Fisher, E. Mammen, J.S. Marron |
| Summary: | Exploratory methods for investigating the structure of a population, such as density estimation, often suggest the presence of modes. These modes can either be spurious artifacts of sampling fluctuation or else be actual features of the underlying distribution. This paper considers hypothesis testing procedures for determining significance of a mode. These procedures are modification of a proposal of Silverman, which is based on kernel density estimation and bootstrap calculation of critical values. |
|---|---|
| Item Description: | Available online 25 March 2002 Gesehen am 26.02.2018 |
| Physical Description: | Online Resource |
| DOI: | 10.1016/0167-9473(94)90080-9 |