REVIEW 6 cited by
DAGnabbit! Ensuring Consistency between Noise and Detection in Hierarchical Bayesian Inference
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Hierarchical Bayesian inference can simultaneously account for both measurement uncertainty and selection effects within astronomical catalogs. In particular, the hierarchy imposed encodes beliefs about the interdependence of the physical processes that generate the observed data. We show that several proposed approximations within the literature actually correspond to inferences that are incompatible with any physical detection process, which can be described by a directed acyclic graph (DAG). This generically leads to biases and is associated with the assumption that detectability is independent of the observed data given the true source parameters. We show several examples of how this error can affect astrophysical inferences based on catalogs of coalescing binaries observed through gravitational waves, including misestimating the redshift evolution of the merger rate as well as incorrectly inferring that General Relativity is the correct theory of gravity when it is not. In general, one cannot directly fit for the ``detected distribution'' and ``divide out'' the selection effects in post-processing. Similarly, when comparing theoretical predictions to observations, it is better to simulate detected data (including both measurement noise and selection effects) rather than comparing estimates of the detected distributions of event parameters (which include only selection effects). While the biases introduced by model misspecification from incorrect assumptions may be smaller than statistical uncertainty for moderate catalog sizes (O(100) events), they will nevertheless pose a significant barrier to precision measurements of astrophysical populations.
Forward citations
Cited by 6 Pith papers
-
When (not) to trust Monte Carlo approximations for hierarchical Bayesian inference
A unified error statistic E-hat measures information lost to Monte Carlo noise in hierarchical Bayesian inference, with a recommended cutoff of 0.2 bits.
-
Gravitational-wave astronomy requires population-informed parameter estimation
Single-event GW parameter estimates under reference priors are population-biased; hierarchical, population-informed re-analysis is needed and changes the identification of the most extreme black holes in the catalog.
-
Functional inference on deviations from General Relativity
GRANITA reconstructs functional, parameter-dependent deviations from General Relativity in gravitational-wave data using Gaussian process regression with free node values.
-
Nowhere left to hide: revealing realistic gravitational-wave populations in high dimensions and high resolution with PixelPop
Modeling all significant correlations with the nonparametric model PixelPop recovers the true black-hole merger rate in a simulated 400-event gravitational-wave catalog, while simpler models introduce bias.
-
Progress toward the detection of the gravitational-wave background from stellar-mass binary black holes: a mock data challenge
A mock data challenge shows that a phase-coherent search for the binary black hole background can recover injected signal fractions in realistic noise, using new treatments of noise uncertainty, finite-duration effect...
-
The Long Road to Alignment: Measuring Black Hole Spin Orientation with Expanding Gravitational-Wave Datasets
Simulated gravitational-wave catalogs show spin-tilt peaks at alignment are hard to confirm even with 1500 events, while integrated tilt fractions are robust.
Discussion (0). Continue with ORCID to comment.