The Shorth Plot
John H. J. Einmahl, Maria Gantner, Günther Sawitzki · Journal of Computational and Graphical Statistics · 2009
The shorth plot is a tool to investigate probability mass concentration. It is a graphical representation of the length of the shorth, the shortest interval covering a certain fraction of the distribution, localized by forcing the intervals considered to contain a given point x. It is easy to compute, avoids bandwidth selection problems, and allows scanning for local as well as for global features of the probability distribution. The good performance of the shorth plot is demonstrated through simulations and real data examples. These data as well as an R-package for computation of the shorth plot are available online.
Cited by 5 works. doi:10.1198/jcgs.2009.08020
Authors
References
- Journal of the Royal Statistical Society.
- Using Kernel Density Estimates to Investigate Multimodality
- Computing and Graphing Highest Density Regions
- The Length of the Shorth
- Excess Mass Estimates and Tests for Multimodality
- Generalized Quantile Processes
- Estimating and Visualizing Conditional Densities
- SiZer for Exploration of Structures in Curves
- The Mode Tree: A Tool for Visualization of Nonparametric Density Features
- Simultaneous Density Estimation of Several Income Distributions