Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts

An important task in the study of fast radio bursts (FRBs) remains the automatic classification of repeating and nonrepeating sources based on their morphological properties. We propose a statistical model that considers a modified logistic regression to classify FRB sources. The classical logistic...

Full description

Saved in:
Bibliographic Details
Main Authors: Antonio Herrera-Martin, Radu V. Craiu, Gwendolyn M. Eadie, David C. Stenning, Derek Bingham, B. M. Gaensler, Ziggy Pleunis, Paul Scholz, Ryan Mckinven, Bikash Kharel, Kiyoshi W. Masui
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astrophysical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-4357/adb623
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849774086999244800
author Antonio Herrera-Martin
Radu V. Craiu
Gwendolyn M. Eadie
David C. Stenning
Derek Bingham
B. M. Gaensler
Ziggy Pleunis
Paul Scholz
Ryan Mckinven
Bikash Kharel
Kiyoshi W. Masui
author_facet Antonio Herrera-Martin
Radu V. Craiu
Gwendolyn M. Eadie
David C. Stenning
Derek Bingham
B. M. Gaensler
Ziggy Pleunis
Paul Scholz
Ryan Mckinven
Bikash Kharel
Kiyoshi W. Masui
author_sort Antonio Herrera-Martin
collection DOAJ
description An important task in the study of fast radio bursts (FRBs) remains the automatic classification of repeating and nonrepeating sources based on their morphological properties. We propose a statistical model that considers a modified logistic regression to classify FRB sources. The classical logistic regression model is modified to accommodate the small proportion of repeaters in the data, a feature that is likely due to the sampling procedure and duration and is not a characteristic of the population of FRB sources. The weighted logistic regression hinges on the choice of a tuning parameter that represents the true proportion τ of repeating FRB sources in the entire population. The proposed method has a sound statistical foundation, direct interpretability, and operates with only five parameters, enabling quicker retraining with added data. Using the CHIME/FRB Collaboration sample of repeating and nonrepeating FRBs and numerical experiments, we achieve a classification accuracy for repeaters of nearly 75% or higher when τ is set in the range of 50%–60%. This implies a tentative high proportion of repeaters, which is surprising, but is also in agreement with recent estimates of τ that are obtained using other methods.
format Article
id doaj-art-6d4b4e52b60842b1a606950f9a44eb66
institution DOAJ
issn 1538-4357
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal
spelling doaj-art-6d4b4e52b60842b1a606950f9a44eb662025-08-20T03:01:51ZengIOP PublishingThe Astrophysical Journal1538-43572025-01-0198214610.3847/1538-4357/adb623Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio BurstsAntonio Herrera-Martin0https://orcid.org/0000-0002-3654-4662Radu V. Craiu1https://orcid.org/0000-0002-1348-8063Gwendolyn M. Eadie2https://orcid.org/0000-0003-3734-8177David C. Stenning3https://orcid.org/0000-0002-9761-4353Derek Bingham4https://orcid.org/0000-0001-5628-7256B. M. Gaensler5https://orcid.org/0000-0002-3382-9558Ziggy Pleunis6https://orcid.org/0000-0002-4795-697XPaul Scholz7https://orcid.org/0000-0002-7374-7119Ryan Mckinven8https://orcid.org/0000-0001-7348-6900Bikash Kharel9https://orcid.org/0009-0008-6166-1095Kiyoshi W. Masui10https://orcid.org/0000-0002-4279-6946David A. Dunlap Department of Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, Canada; Department of Statistical Science, University of Toronto , Ontario Power Building, 700 University Avenue, 9th Floor, Toronto, ON M5G 1Z5, CanadaDepartment of Statistical Science, University of Toronto , Ontario Power Building, 700 University Avenue, 9th Floor, Toronto, ON M5G 1Z5, CanadaDavid A. Dunlap Department of Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, Canada; Department of Statistical Science, University of Toronto , Ontario Power Building, 700 University Avenue, 9th Floor, Toronto, ON M5G 1Z5, CanadaDepartment of Statistics and Actuarial Science, Simon Fraser University , 8888 University Drive, Burnaby, BC V5A 1S6, CanadaDepartment of Statistics and Actuarial Science, Simon Fraser University , 8888 University Drive, Burnaby, BC V5A 1S6, CanadaDavid A. Dunlap Department of Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, Canada; Department of Astronomy and Astrophysics, University of California Santa Cruz , 1156 High Street, Santa Cruz, CA 95064, USA; Dunlap Institute for Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, CanadaDunlap Institute for Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, Canada; Anton Pannekoek Institute for Astronomy, University of Amsterdam , Science Park 904, 1098 XH Amsterdam, The Netherlands; ASTRON , Netherlands Institute for Radio Astronomy, Oude Hoogeveensedijk 4, 7991 PD Dwingeloo, The NetherlandsDunlap Institute for Astronomy & Astrophysics, University of Toronto , 50 St. George Street, Toronto, ON M5S 3H4, Canada; Department of Physics and Astronomy, York University , 4700 Keele Street, Toronto, ON MJ3 1P3, CanadaDepartment of Physics, McGill University , 3600 rue University, Montréal, QC H3A 2T8, Canada; Trottier Space Institute, McGill University , 3550 rue University, Montréal, QC H3A 2A7, CanadaDepartment of Physics and Astronomy, West Virginia University , P.O. Box 6315, Morgantown, WV 26506, USAMIT Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of Technology , 77 Massachusetts Ave, Cambridge, MA 02139, USA; Department of Physics, Massachusetts Institute of Technology , 77 Massachusetts Ave, Cambridge, MA 02139, USAAn important task in the study of fast radio bursts (FRBs) remains the automatic classification of repeating and nonrepeating sources based on their morphological properties. We propose a statistical model that considers a modified logistic regression to classify FRB sources. The classical logistic regression model is modified to accommodate the small proportion of repeaters in the data, a feature that is likely due to the sampling procedure and duration and is not a characteristic of the population of FRB sources. The weighted logistic regression hinges on the choice of a tuning parameter that represents the true proportion τ of repeating FRB sources in the entire population. The proposed method has a sound statistical foundation, direct interpretability, and operates with only five parameters, enabling quicker retraining with added data. Using the CHIME/FRB Collaboration sample of repeating and nonrepeating FRBs and numerical experiments, we achieve a classification accuracy for repeaters of nearly 75% or higher when τ is set in the range of 50%–60%. This implies a tentative high proportion of repeaters, which is surprising, but is also in agreement with recent estimates of τ that are obtained using other methods.https://doi.org/10.3847/1538-4357/adb623Radio transient sourcesClassificationAstrostatistics techniquesSampling distributionAstrostatistics distributionsRegression
spellingShingle Antonio Herrera-Martin
Radu V. Craiu
Gwendolyn M. Eadie
David C. Stenning
Derek Bingham
B. M. Gaensler
Ziggy Pleunis
Paul Scholz
Ryan Mckinven
Bikash Kharel
Kiyoshi W. Masui
Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
The Astrophysical Journal
Radio transient sources
Classification
Astrostatistics techniques
Sampling distribution
Astrostatistics distributions
Regression
title Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
title_full Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
title_fullStr Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
title_full_unstemmed Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
title_short Rare Event Classification with Weighted Logistic Regression for Identifying Repeating Fast Radio Bursts
title_sort rare event classification with weighted logistic regression for identifying repeating fast radio bursts
topic Radio transient sources
Classification
Astrostatistics techniques
Sampling distribution
Astrostatistics distributions
Regression
url https://doi.org/10.3847/1538-4357/adb623
work_keys_str_mv AT antonioherreramartin rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT raduvcraiu rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT gwendolynmeadie rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT davidcstenning rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT derekbingham rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT bmgaensler rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT ziggypleunis rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT paulscholz rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT ryanmckinven rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT bikashkharel rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts
AT kiyoshiwmasui rareeventclassificationwithweightedlogisticregressionforidentifyingrepeatingfastradiobursts