Autoregressive Planet Search: Methodology

Gabriel A. Caceres, Eric Feigelson, G. Jogesh Babu, Natalia Bahamonde, Alejandra Christen, Karine Bertin, Cristian Meza, Michel Curé

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The detection of periodic signals from transiting exoplanets is often impeded by extraneous aperiodic photometric variability, either intrinsic to the star or arising from the measurement process. Frequently, these variations are autocorrelated wherein later flux values are correlated with previous ones. In this work, we present the methodology of the autoregessive planet search (ARPS) project, which uses the autoregressive integrated moving average (ARIMA) and related statistical models that treat a wide variety of stochastic processes, as well as nonstationarity, to improve detection of new planetary transits. Provided a time series is evenly spaced or can be placed on an evenly spaced grid with missing values, these low-dimensional parametric models can prove very effective. We introduce a planet search algorithm to detect periodic transits in the residuals after the application of ARIMA models. Our matched-filter algorithm, the transit comb filter (TCF), is closely related to the traditional box-fitting least-squares and provides an analogous periodogram. Finally, if a previously identified or simulated sample of planets is available, selected scalar features from different stages of the analysis - the original light curves, ARIMA fits, TCF periodograms, and folded light curves - can be collectively used with a multivariate classifier to identify promising candidates while efficiently rejecting false alarms. We use Random Forests for this task, in conjunction with receiver operating characteristic curves, to define discovery criteria for new, high-fidelity planetary candidates. The ARPS methodology can be applied to both evenly spaced satellite light curves and densely cadenced ground-based photometric surveys.

Original languageEnglish (US)
Article number57
JournalAstronomical Journal
Volume158
Issue number2
DOIs
StatePublished - Jan 1 2019

Fingerprint

transit
planets
planet
methodology
light curve
filter
filters
matched filters
false alarms
stochastic processes
extrasolar planets
stochasticity
classifiers
boxes
receivers
grids
time series
scalars
stars
curves

All Science Journal Classification (ASJC) codes

  • Astronomy and Astrophysics
  • Space and Planetary Science

Cite this

Caceres, G. A., Feigelson, E., Babu, G. J., Bahamonde, N., Christen, A., Bertin, K., ... Curé, M. (2019). Autoregressive Planet Search: Methodology. Astronomical Journal, 158(2), [57]. https://doi.org/10.3847/1538-3881/ab26b8
Caceres, Gabriel A. ; Feigelson, Eric ; Babu, G. Jogesh ; Bahamonde, Natalia ; Christen, Alejandra ; Bertin, Karine ; Meza, Cristian ; Curé, Michel. / Autoregressive Planet Search : Methodology. In: Astronomical Journal. 2019 ; Vol. 158, No. 2.
@article{3e7b815af8eb42c3bc306beca604a920,
title = "Autoregressive Planet Search: Methodology",
abstract = "The detection of periodic signals from transiting exoplanets is often impeded by extraneous aperiodic photometric variability, either intrinsic to the star or arising from the measurement process. Frequently, these variations are autocorrelated wherein later flux values are correlated with previous ones. In this work, we present the methodology of the autoregessive planet search (ARPS) project, which uses the autoregressive integrated moving average (ARIMA) and related statistical models that treat a wide variety of stochastic processes, as well as nonstationarity, to improve detection of new planetary transits. Provided a time series is evenly spaced or can be placed on an evenly spaced grid with missing values, these low-dimensional parametric models can prove very effective. We introduce a planet search algorithm to detect periodic transits in the residuals after the application of ARIMA models. Our matched-filter algorithm, the transit comb filter (TCF), is closely related to the traditional box-fitting least-squares and provides an analogous periodogram. Finally, if a previously identified or simulated sample of planets is available, selected scalar features from different stages of the analysis - the original light curves, ARIMA fits, TCF periodograms, and folded light curves - can be collectively used with a multivariate classifier to identify promising candidates while efficiently rejecting false alarms. We use Random Forests for this task, in conjunction with receiver operating characteristic curves, to define discovery criteria for new, high-fidelity planetary candidates. The ARPS methodology can be applied to both evenly spaced satellite light curves and densely cadenced ground-based photometric surveys.",
author = "Caceres, {Gabriel A.} and Eric Feigelson and Babu, {G. Jogesh} and Natalia Bahamonde and Alejandra Christen and Karine Bertin and Cristian Meza and Michel Cur{\'e}",
year = "2019",
month = "1",
day = "1",
doi = "10.3847/1538-3881/ab26b8",
language = "English (US)",
volume = "158",
journal = "Astronomical Journal",
issn = "0004-6256",
publisher = "IOP Publishing Ltd.",
number = "2",

}

Caceres, GA, Feigelson, E, Babu, GJ, Bahamonde, N, Christen, A, Bertin, K, Meza, C & Curé, M 2019, 'Autoregressive Planet Search: Methodology', Astronomical Journal, vol. 158, no. 2, 57. https://doi.org/10.3847/1538-3881/ab26b8

Autoregressive Planet Search : Methodology. / Caceres, Gabriel A.; Feigelson, Eric; Babu, G. Jogesh; Bahamonde, Natalia; Christen, Alejandra; Bertin, Karine; Meza, Cristian; Curé, Michel.

In: Astronomical Journal, Vol. 158, No. 2, 57, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Autoregressive Planet Search

T2 - Methodology

AU - Caceres, Gabriel A.

AU - Feigelson, Eric

AU - Babu, G. Jogesh

AU - Bahamonde, Natalia

AU - Christen, Alejandra

AU - Bertin, Karine

AU - Meza, Cristian

AU - Curé, Michel

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The detection of periodic signals from transiting exoplanets is often impeded by extraneous aperiodic photometric variability, either intrinsic to the star or arising from the measurement process. Frequently, these variations are autocorrelated wherein later flux values are correlated with previous ones. In this work, we present the methodology of the autoregessive planet search (ARPS) project, which uses the autoregressive integrated moving average (ARIMA) and related statistical models that treat a wide variety of stochastic processes, as well as nonstationarity, to improve detection of new planetary transits. Provided a time series is evenly spaced or can be placed on an evenly spaced grid with missing values, these low-dimensional parametric models can prove very effective. We introduce a planet search algorithm to detect periodic transits in the residuals after the application of ARIMA models. Our matched-filter algorithm, the transit comb filter (TCF), is closely related to the traditional box-fitting least-squares and provides an analogous periodogram. Finally, if a previously identified or simulated sample of planets is available, selected scalar features from different stages of the analysis - the original light curves, ARIMA fits, TCF periodograms, and folded light curves - can be collectively used with a multivariate classifier to identify promising candidates while efficiently rejecting false alarms. We use Random Forests for this task, in conjunction with receiver operating characteristic curves, to define discovery criteria for new, high-fidelity planetary candidates. The ARPS methodology can be applied to both evenly spaced satellite light curves and densely cadenced ground-based photometric surveys.

AB - The detection of periodic signals from transiting exoplanets is often impeded by extraneous aperiodic photometric variability, either intrinsic to the star or arising from the measurement process. Frequently, these variations are autocorrelated wherein later flux values are correlated with previous ones. In this work, we present the methodology of the autoregessive planet search (ARPS) project, which uses the autoregressive integrated moving average (ARIMA) and related statistical models that treat a wide variety of stochastic processes, as well as nonstationarity, to improve detection of new planetary transits. Provided a time series is evenly spaced or can be placed on an evenly spaced grid with missing values, these low-dimensional parametric models can prove very effective. We introduce a planet search algorithm to detect periodic transits in the residuals after the application of ARIMA models. Our matched-filter algorithm, the transit comb filter (TCF), is closely related to the traditional box-fitting least-squares and provides an analogous periodogram. Finally, if a previously identified or simulated sample of planets is available, selected scalar features from different stages of the analysis - the original light curves, ARIMA fits, TCF periodograms, and folded light curves - can be collectively used with a multivariate classifier to identify promising candidates while efficiently rejecting false alarms. We use Random Forests for this task, in conjunction with receiver operating characteristic curves, to define discovery criteria for new, high-fidelity planetary candidates. The ARPS methodology can be applied to both evenly spaced satellite light curves and densely cadenced ground-based photometric surveys.

UR - http://www.scopus.com/inward/record.url?scp=85072024169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072024169&partnerID=8YFLogxK

U2 - 10.3847/1538-3881/ab26b8

DO - 10.3847/1538-3881/ab26b8

M3 - Article

AN - SCOPUS:85072024169

VL - 158

JO - Astronomical Journal

JF - Astronomical Journal

SN - 0004-6256

IS - 2

M1 - 57

ER -

Caceres GA, Feigelson E, Babu GJ, Bahamonde N, Christen A, Bertin K et al. Autoregressive Planet Search: Methodology. Astronomical Journal. 2019 Jan 1;158(2). 57. https://doi.org/10.3847/1538-3881/ab26b8