Coping with Selection Effects: A Primer on Regression with Truncated Data
arXiv:1901.10522 · doi:10.1093/mnras/stz320
Abstract
The finite sensitivity of instruments or detection methods means that data sets in many areas of astronomy, for example cosmological or exoplanet surveys, are necessarily systematically incomplete. Such data sets, where the population being investigated is of unknown size and only partially represented in the data, are called "truncated" in the statistical literature. Truncation can be accounted for through a relatively straightforward modification to the model being fitted in many circumstances, provided that the model can be extended to describe the population of undetected sources. Here I examine the problem of regression using truncated data in general terms, and use a simple example to show the impact of selecting a subset of potential data on the dependent variable, on the independent variable, and on a second dependent variable that is correlated with the variable of interest. Special circumstances in which selection effects are ignorable are noted. I also comment on computational strategies for performing regression with truncated data, as an extension of methods that have become popular for the non-truncated case, and provide some general recommendations.
arXiv comment11 pages, to appear in MNRAS
References in corpus (18)
- Some Aspects of Measurement Error in Linear Regression of Astronomical Data
- Cosmological Parameters from Observations of Galaxy Clusters
- Chandra Cluster Cosmology Project II: Samples and X-ray Data Reduction
- Galaxy cluster X-ray luminosity scaling relations from a representative local sample (REXCESS)
- Extracting distribution parameters from multiple uncertain observations with selection biases
- The Observed Growth of Massive Galaxy Clusters I: Statistical Methods and Cosmological Constraints
- On the Accuracy of Weak Lensing Cluster Mass Reconstructions
- The Observed Growth of Massive Galaxy Clusters II: X-ray Scaling Relations
- The Exoplanet Census: A General Method, Applied to Kepler
- The XMM-LSS survey: the Class 1 cluster sample over the initial 5 square degrees and its cosmological modelling
- Scaling relations for galaxy clusters: properties and evolution
- Cosmological Evolution of Long Gamma-ray Bursts and Star Formation Rate
- A Model for Multi-property Galaxy Cluster Statistics
- Objects in Kepler's Mirror May be Larger Than They Appear: Bias and Selection Effects in Transiting Planet Surveys
- CoMaLit - II. The scaling relation between mass and Sunyaev-Zel'dovich signal for Planck selected galaxy clusters
- Richness-based masses of rich and famous galaxy clusters
- A Gibbs Sampler for Multivariate Linear Regression
- A Bayesian approach to truncated data sets: An application to Malmquist bias in Supernova Cosmology
Cited by in corpus (12)
- CCCP and MENeaCS: (updated) weak-lensing masses for 100 galaxy clusters
- Dissecting the size-mass and -mass relations at 1.0 < z < 2.5: galaxy mass profiles and color gradients as a function of spectral shape
- The eROSITA Final Equatorial-Depth Survey (eFEDS): X-ray Properties and Scaling Relations of Galaxy Clusters and Groups
- Cosmology dependency of halo masses and concentrations in hydrodynamic simulations
- How baryons can significantly bias cluster count cosmology
- Mass Variance from Archival X-ray Properties of Dark Energy Survey Year-1 Galaxy Clusters
- Effects of Multiphase Gas and Projection on X-ray Observables in Simulated Galaxy Clusters as Seen by eROSITA
- Ellipticity of Brightest Cluster Galaxies as tracer of halo orientation and weak-lensing mass bias
- Why are we still using 3D masses for cluster cosmology?
- X-ray cluster cosmology
- PoPE: A population-based approach to model spatial structure of astronomical systems
- HSC-XXL : Baryon budget of the 136 XXL Groups and Clusters