Excess Mass Estimates and Tests for Multimodality
- 1 September 1991
- journal article
- theory and-methods
- Published by Taylor & Francis in Journal of the American Statistical Association
- Vol. 86 (415) , 738-746
- https://doi.org/10.1080/01621459.1991.10475103
Abstract
We propose and study a method for analyzing the modality of a distribution. The method is based on the excess mass functional. This functional measures excessive empirical mass in comparison with multiples of uniform distribution. By the excess mass approach we separate the investigation about the number of modes from questions concerning their location. For distributions on the line, the excess mass functional can be estimated at a square-root rate. The asymptotic behavior of estimators is studied, and tests for multimodality based on the excess mass are derived.Keywords
This publication has 8 references indexed in Scilit:
- One-Sided Inference about Functionals of a DensityThe Annals of Statistics, 1988
- Estimation of a Convex Density Contour in Two DimensionsJournal of the American Statistical Association, 1987
- A bootstrap testing procedure for investigating the number of subpopulationsJournal of Statistical Computation and Simulation, 1985
- The Dip Test of UnimodalityThe Annals of Statistics, 1985
- Using Kernel Density Estimates to Investigate MultimodalityJournal of the Royal Statistical Society Series B: Statistical Methodology, 1981
- Density Estimation and Bump-Hunting by the Penalized Likelihood Method Exemplified by Scattering and Meteorite DataJournal of the American Statistical Association, 1980
- An approximation of partial sums of independent RV'-s, and the sample DF. IProbability Theory and Related Fields, 1975
- Observations on the Fe-Si-Mg relationship in chondritesGeochimica et Cosmochimica Acta, 1965