Selection Functions in Astronomical Data Modeling, with the Space Density of White Dwarfs as a Worked Example
Journal
Date Issued
2021
Author(s)
Rix, Hans-Walter
•
Hogg, David W.
•
Boubert, Douglas
•
Brown, Anthony G. A.
•
Casey, Andrew
•
•
Everall, Andrew
•
Fouesneau, Morgan
•
Price-Whelan, Adrian M.
Abstract
Statistical studies of cataloged object properties are central to astrophysics. But one cannot model those objects' population properties without the sample's selection function, the quantitative understanding of which objects could have ended up in such a catalog. As didactic introductions to this topic are scarce in the astrophysical literature, we provide one here, addressing the following questions: What is a selection function? On what arguments q should it depend? Over what domain must a selection function be defined? What simplifying approximations can be made? And, how is a selection function used in "modeling"? We argue that volume-complete samples, limited by the faintest objects, reflect a highly suboptimal selection function, needlessly reducing the number of bright and usually rare sample members. We illustrate these points by a worked example: github.com/gaia-unlimited/WD-selection-function, deriving the space density of white dwarfs (WDs) in the Galactic neighborhood as a function of their luminosity and color, Φ0(MG, (B - R)) in [mag-2 pc-3]. We construct a sample ${ \mathcal C }$ of 105 presumed WDs through straightforward selection cuts on the Gaia EDR3 catalog in magnitude, color, and parallax, q = (G, (B - R), ϖ). We then combine a simple model for Φ0 with this selection function's ${S}_{{ \mathcal C }}({\boldsymbol{q}})$ effective survey volume to estimate Φ0(MG, (B - R)) precisely and robustly against the detailed choices for ${S}_{{ \mathcal C }}({\boldsymbol{q}})$ . This resulting WD luminosity-color function Φ0(MG, (B - R)) differs dramatically from the initial number density distribution in the luminosity-color plane: by orders of magnitude in density and by four magnitudes in density peak location.
Volume
162
Issue
4
Start page
142
Issn Identifier
0004-6256
Ads BibCode
2021AJ....162..142R
Rights
open.access
File(s)![Thumbnail Image]()
Loading...
Name
Rix_2021_AJ_162_142.pdf
Description
PDF editoriale
Size
1.55 MB
Format
Adobe PDF
Checksum (MD5)
81fc6f833d2c80b6813aa698363b90dc