• Corpus ID: 7428254

Error Probabilities for Halfspace Depth

@article{Burr2016ErrorPF,
  title={Error Probabilities for Halfspace Depth},
  author={Michael A. Burr and Robert Fabrizio},
  journal={ArXiv},
  year={2016},
  volume={abs/1605.04323}
}
Data depth functions are a generalization of one-dimensional order statistics and medians to real spaces of dimension greater than one; in particular, a data depth function quantifies the centrality of a point with respect to a data set or a probability distribution. One of the most commonly studied data depth functions is halfspace depth. It is of interest to computational geometers because it is highly geometric, and it is of interest to statisticians because it shares many desirable… 
1 Citations

Figures from this paper

Generalization of the simplicial depth: no vanishment outside the convex hull of the distribution support
The simplicial depth, like other relevant multivariate statistical data depth functions, vanishes right outside the convex hull of the support of the distribution with respect to which the depth is

References

SHOWING 1-10 OF 30 REFERENCES
Dynamic Maintenance of Half-Space Depth for Points and Contours
TLDR
A dynamic algorithm for dynamically maintaining the rank-based contours in $O(n\cdot\log n)$ time per update and overall quadratic space and a strong structural result of independent interest describing the behavior of dynamic cover- based contours near data points are presented.
General notions of statistical depth function
Statistical depth functions are being formulated ad hoc with increasing popularity in nonparametric inference for multivariate data. Here we introduce several general structures for depth functions,
Quantitative asymptotics of graphical projection pursuit
There is a result of Diaconis and Freedman which says that, in a limiting sense, for large collections of high-dimensional data most one-dimensional projections of the data are approximately
Covering the Sphere by Equal Spherical Balls
We show that for any acute ϕ, there exists a covering of S d by spherical balls of radius ϕ such that no point is covered more than 400d ln d times. It follows that the density is of order at most d
Empirical Processes with Applications to Statistics (wiley Series in Probability and Mathematical Statistics)
type, all depending on an additional parameter which assumes a particular value for the problem in question. The relationship between this idea and dynamic programming, which is a technique for
Upper and lower class sequences for minimal uniform spacings
SummaryLetKn be thek-th smallest spacing defined byX1,...,Xn, independent random variables uniformly distributed on [0,1], wherek≧1 is a fixed integer. Letun be a sequence of positive numbers
Geometric measures of data depth
  • G. Aloupis
  • Computer Science
    Data Depth: Robust Multivariate Analysis, Computational Geometry and Applications
  • 2003
Several measures of data depth have been proposed, each attempting to maintain certain robustness properties. This paper lists the main approaches known to the computer science community. Properties
Empirical Processes with Applications to Statistics
Here is the first book to summarize a broad cross-section of the large volume of literature available on one-dimensional empirical processes. Presented is a thorough treatment of the theory of
Combinatorial methods in density estimation
TLDR
A comparison of the Kernel Estimate and the Vapnik-Chervonenkis Dimension and Covering Numbers shows that the former is significantly more accurate than the latter and the latter is significantly less accurate.
...
1
2
3
...