| |
 |
| Paper: |
INDICATE: A Tool to Quantify the Degree of Association of Each Point in a 2+D Discrete Dataset |
| Volume: |
538, ADASS XXXII |
| Page: |
234 |
| Authors: |
Anne S. M. Buckner |
| DOI: |
10.26624/LOFL4398 |
| Abstract: |
INDICATE is a novel statistical clustering tool to quantify the degree of
spatial clustering of points in a discrete 2+D dataset that is available to download as
a Python software package. It employs a nearest-neighbour approach to quantitatively
compare the local spatial distribution in a point’s neighbourhood with that expected if
the point were located in non-spatially clustered region. An index, I, is subsequently
assigned to each point denoting its degree of spatial association. This index is calibrated
against random distributions to aid interpretation and differentiation between points in
spatially random and clustered distributions, and is independent of dataset shape, size
or density. As such, results obtained for different datasets are directly comparable.
INDICATE is an ideal choice for astronomical data analysis as it can be applied to both
observed and simulated datasets, consisting of any object type (e.g., stars, galaxies,
sinks, particles) or parameter space (e.g., position, colour, magnitude) and is robust
against dataset incompleteness up to 83.3%, outliers and edge effects. |
|
|
|
|
|
