Model selection and localisation biases for repeating FRBs

Determine how biases introduced by the follow-up observations required to detect at least one repeat burst affect the use of repeating fast radio bursts and their host-galaxy information in population analyses.

Background

The framework could incorporate the number of observed repeats as a radio property and use repeating FRB redshifts to constrain the redshift distribution and compare host-galaxy populations of repeaters and apparently once-off bursts. However, most repeaters must first be detected in follow-up observations before they can be localised precisely enough for host identification.

This creates a selection effect: the probability of obtaining a localisation and host association may depend on the repeat rate, burst detectability, and follow-up strategy. The paper does not yet specify how to quantify or correct these effects, leaving their impact on inferred repeating-FRB populations unresolved.

References

There is one complication to this optimism however. In order to localise most repeaters, follow-up observations need to detect at least one burst. Care will therefore have to be taken to account for any such biases arising from the follow-up procedure. We also leave such considerations to a future work.

— Through a glass, darkly: a combined framework for estimating fast radio burst host galaxy and population properties in an era of uncertain host identification  (2609.04765 - James et al., 4 Sep 2026) in Section 7.2, “Inclusion of repeating FRBs”