2011
DOI: 10.1051/0004-6361/201116878
|View full text |Cite
|
Sign up to set email alerts
|

Comparison of density estimation methods for astronomical datasets

Abstract: Context. Galaxies are strongly influenced by their environment. Quantifying the galaxy density is a difficult but critical step in studying the properties of galaxies. Aims. We aim to determine differences in density estimation methods and their applicability in astronomical problems. We study the performance of four density estimation techniques: k-nearest neighbors (kNN), adaptive Gaussian kernel density estimation (DEDICA), a special case of adaptive Epanechnikov kernel density estimation (MBE), and the Del… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
34
0

Year Published

2012
2012
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 34 publications
(34 citation statements)
references
References 44 publications
0
34
0
Order By: Relevance
“…We use a modified Breiman density estimator (Ferdosi et al 2011) to infer f a and f a shuf from the set of stellar actions, then calculate the KLD using numerical integration over a regular grid of J, which replaces the integral in Equation (12) with a sum over grid squares. More details on the numerical methods are available in Sanderson et al (2015).…”
Section: Measuring Clustering With the Kldmentioning
confidence: 99%
“…We use a modified Breiman density estimator (Ferdosi et al 2011) to infer f a and f a shuf from the set of stellar actions, then calculate the KLD using numerical integration over a regular grid of J, which replaces the integral in Equation (12) with a sum over grid squares. More details on the numerical methods are available in Sanderson et al (2015).…”
Section: Measuring Clustering With the Kldmentioning
confidence: 99%
“…In many papers (e.g. Vio et al 1994;Fadda et al 1998;Ferdosi et al 2011) it has been shown that kernel smoothing is the best and recommended choice for density estimation. It is more robust and reliable than other simpler methods and provides comparable (or better) results than methods with high computational costs (e.g.…”
Section: Appendix A: Kernel Density Estimationmentioning
confidence: 99%
“…Such methods effectively 'smear' each particle over a region and assign to each point in space a scalar value that approximates the local particle density. In our case, we compute a value for the scalar density field at each grid node using the modified Breiman kernel density estimation method (MBE) with a finite-support adaptive Epanechnikov kernel [9,35]. Our reason for choosing this particular method is that it was shown by Ferdosi et al [9] to be optimal with respect to speed and reliability when compared to the k-nearest neighbors, adaptive Gaussian kernel, and Delaunay tessellation field methods.…”
Section: Density Estimationmentioning
confidence: 99%
“…We now give a brief description of the MBE method and we refer to Ferdosi et al's paper [9] for more details. First, for each direction k = x, y, z we compute a smoothing length as…”
Section: Density Estimationmentioning
confidence: 99%