Photometric redshifts and quasar probabilities from a single, data-driven generative model

Jo Bovy, Adam D. Myers, Joseph F. Hennawi, David W. Hogg, Richard G. McMahon, David Schiminovich, Erin S. Sheldon, Jon Brinkmann, Donald P. Schneider, Benjamin A. Weaver

    Research output: Contribution to journalArticle

    Abstract

    We describe a technique for simultaneously classifying and estimating the redshift of quasars. It can separate quasars from stars in arbitrary redshift ranges, estimate full posterior distribution functions for the redshift, and naturally incorporate flux uncertainties, missing data, and multi-wavelength photometry. We build models of quasars in flux-redshift space by applying the extreme deconvolution technique to estimate the underlying density. By integrating this density over redshift, one can obtain quasar flux densities in different redshift ranges. This approach allows for efficient, consistent, and fast classification and photometric redshift estimation. This is achieved by combining the speed obtained by choosing simple analytical forms as the basis of our density model with the flexibility of non-parametric models through the use of many simple components with many parameters. We show that this technique is competitive with the best photometric quasar classification techniques - which are limited to fixed, broad redshift ranges and high signal-to-noise ratio data - and with the best photometric redshift techniques when applied to broadband optical data. We demonstrate that the inclusion of UV and NIR data significantly improves photometric quasar-star separation and essentially resolves all of the redshift degeneracies for quasars inherent to the ugriz filter system, even when included data have a low signal-to-noise ratio. For quasars spectroscopically confirmed by the SDSS 84% and 97% of the objects with Galaxy Evolution Explorer UV and UKIDSS NIR data have photometric redshifts within 0.1 and 0.3, respectively, of the spectroscopic redshift; this amounts to about a factor of three improvement over ugriz-only photometric redshifts. Our code to calculate quasar probabilities and redshift probability distributions is publicly available.

    Original languageEnglish (US)
    Article number41
    JournalAstrophysical Journal
    Volume749
    Issue number1
    DOIs
    StatePublished - Apr 10 2012

    Fingerprint

    quasars
    signal-to-noise ratio
    signal to noise ratios
    deconvolution
    stars
    estimates
    filter
    classifying
    wavelength
    photometry
    flexibility
    estimating
    flux density
    distribution functions
    inclusions
    galaxies
    broadband
    filters
    wavelengths
    distribution

    Keywords

    • catalogs
    • cosmology: observations
    • galaxies: distances and redshifts
    • galaxies: photometry
    • methods: data analysis
    • quasars: general

    ASJC Scopus subject areas

    • Space and Planetary Science
    • Astronomy and Astrophysics

    Cite this

    Bovy, J., Myers, A. D., Hennawi, J. F., Hogg, D. W., McMahon, R. G., Schiminovich, D., ... Weaver, B. A. (2012). Photometric redshifts and quasar probabilities from a single, data-driven generative model. Astrophysical Journal, 749(1), [41]. https://doi.org/10.1088/0004-637X/749/1/41

    Photometric redshifts and quasar probabilities from a single, data-driven generative model. / Bovy, Jo; Myers, Adam D.; Hennawi, Joseph F.; Hogg, David W.; McMahon, Richard G.; Schiminovich, David; Sheldon, Erin S.; Brinkmann, Jon; Schneider, Donald P.; Weaver, Benjamin A.

    In: Astrophysical Journal, Vol. 749, No. 1, 41, 10.04.2012.

    Research output: Contribution to journalArticle

    Bovy, J, Myers, AD, Hennawi, JF, Hogg, DW, McMahon, RG, Schiminovich, D, Sheldon, ES, Brinkmann, J, Schneider, DP & Weaver, BA 2012, 'Photometric redshifts and quasar probabilities from a single, data-driven generative model', Astrophysical Journal, vol. 749, no. 1, 41. https://doi.org/10.1088/0004-637X/749/1/41
    Bovy, Jo ; Myers, Adam D. ; Hennawi, Joseph F. ; Hogg, David W. ; McMahon, Richard G. ; Schiminovich, David ; Sheldon, Erin S. ; Brinkmann, Jon ; Schneider, Donald P. ; Weaver, Benjamin A. / Photometric redshifts and quasar probabilities from a single, data-driven generative model. In: Astrophysical Journal. 2012 ; Vol. 749, No. 1.
    @article{c2a2bda6f73346ba91d1bd682adc5d20,
    title = "Photometric redshifts and quasar probabilities from a single, data-driven generative model",
    abstract = "We describe a technique for simultaneously classifying and estimating the redshift of quasars. It can separate quasars from stars in arbitrary redshift ranges, estimate full posterior distribution functions for the redshift, and naturally incorporate flux uncertainties, missing data, and multi-wavelength photometry. We build models of quasars in flux-redshift space by applying the extreme deconvolution technique to estimate the underlying density. By integrating this density over redshift, one can obtain quasar flux densities in different redshift ranges. This approach allows for efficient, consistent, and fast classification and photometric redshift estimation. This is achieved by combining the speed obtained by choosing simple analytical forms as the basis of our density model with the flexibility of non-parametric models through the use of many simple components with many parameters. We show that this technique is competitive with the best photometric quasar classification techniques - which are limited to fixed, broad redshift ranges and high signal-to-noise ratio data - and with the best photometric redshift techniques when applied to broadband optical data. We demonstrate that the inclusion of UV and NIR data significantly improves photometric quasar-star separation and essentially resolves all of the redshift degeneracies for quasars inherent to the ugriz filter system, even when included data have a low signal-to-noise ratio. For quasars spectroscopically confirmed by the SDSS 84{\%} and 97{\%} of the objects with Galaxy Evolution Explorer UV and UKIDSS NIR data have photometric redshifts within 0.1 and 0.3, respectively, of the spectroscopic redshift; this amounts to about a factor of three improvement over ugriz-only photometric redshifts. Our code to calculate quasar probabilities and redshift probability distributions is publicly available.",
    keywords = "catalogs, cosmology: observations, galaxies: distances and redshifts, galaxies: photometry, methods: data analysis, quasars: general",
    author = "Jo Bovy and Myers, {Adam D.} and Hennawi, {Joseph F.} and Hogg, {David W.} and McMahon, {Richard G.} and David Schiminovich and Sheldon, {Erin S.} and Jon Brinkmann and Schneider, {Donald P.} and Weaver, {Benjamin A.}",
    year = "2012",
    month = "4",
    day = "10",
    doi = "10.1088/0004-637X/749/1/41",
    language = "English (US)",
    volume = "749",
    journal = "Astrophysical Journal",
    issn = "0004-637X",
    publisher = "IOP Publishing Ltd.",
    number = "1",

    }

    TY - JOUR

    T1 - Photometric redshifts and quasar probabilities from a single, data-driven generative model

    AU - Bovy, Jo

    AU - Myers, Adam D.

    AU - Hennawi, Joseph F.

    AU - Hogg, David W.

    AU - McMahon, Richard G.

    AU - Schiminovich, David

    AU - Sheldon, Erin S.

    AU - Brinkmann, Jon

    AU - Schneider, Donald P.

    AU - Weaver, Benjamin A.

    PY - 2012/4/10

    Y1 - 2012/4/10

    N2 - We describe a technique for simultaneously classifying and estimating the redshift of quasars. It can separate quasars from stars in arbitrary redshift ranges, estimate full posterior distribution functions for the redshift, and naturally incorporate flux uncertainties, missing data, and multi-wavelength photometry. We build models of quasars in flux-redshift space by applying the extreme deconvolution technique to estimate the underlying density. By integrating this density over redshift, one can obtain quasar flux densities in different redshift ranges. This approach allows for efficient, consistent, and fast classification and photometric redshift estimation. This is achieved by combining the speed obtained by choosing simple analytical forms as the basis of our density model with the flexibility of non-parametric models through the use of many simple components with many parameters. We show that this technique is competitive with the best photometric quasar classification techniques - which are limited to fixed, broad redshift ranges and high signal-to-noise ratio data - and with the best photometric redshift techniques when applied to broadband optical data. We demonstrate that the inclusion of UV and NIR data significantly improves photometric quasar-star separation and essentially resolves all of the redshift degeneracies for quasars inherent to the ugriz filter system, even when included data have a low signal-to-noise ratio. For quasars spectroscopically confirmed by the SDSS 84% and 97% of the objects with Galaxy Evolution Explorer UV and UKIDSS NIR data have photometric redshifts within 0.1 and 0.3, respectively, of the spectroscopic redshift; this amounts to about a factor of three improvement over ugriz-only photometric redshifts. Our code to calculate quasar probabilities and redshift probability distributions is publicly available.

    AB - We describe a technique for simultaneously classifying and estimating the redshift of quasars. It can separate quasars from stars in arbitrary redshift ranges, estimate full posterior distribution functions for the redshift, and naturally incorporate flux uncertainties, missing data, and multi-wavelength photometry. We build models of quasars in flux-redshift space by applying the extreme deconvolution technique to estimate the underlying density. By integrating this density over redshift, one can obtain quasar flux densities in different redshift ranges. This approach allows for efficient, consistent, and fast classification and photometric redshift estimation. This is achieved by combining the speed obtained by choosing simple analytical forms as the basis of our density model with the flexibility of non-parametric models through the use of many simple components with many parameters. We show that this technique is competitive with the best photometric quasar classification techniques - which are limited to fixed, broad redshift ranges and high signal-to-noise ratio data - and with the best photometric redshift techniques when applied to broadband optical data. We demonstrate that the inclusion of UV and NIR data significantly improves photometric quasar-star separation and essentially resolves all of the redshift degeneracies for quasars inherent to the ugriz filter system, even when included data have a low signal-to-noise ratio. For quasars spectroscopically confirmed by the SDSS 84% and 97% of the objects with Galaxy Evolution Explorer UV and UKIDSS NIR data have photometric redshifts within 0.1 and 0.3, respectively, of the spectroscopic redshift; this amounts to about a factor of three improvement over ugriz-only photometric redshifts. Our code to calculate quasar probabilities and redshift probability distributions is publicly available.

    KW - catalogs

    KW - cosmology: observations

    KW - galaxies: distances and redshifts

    KW - galaxies: photometry

    KW - methods: data analysis

    KW - quasars: general

    UR - http://www.scopus.com/inward/record.url?scp=84858976239&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84858976239&partnerID=8YFLogxK

    U2 - 10.1088/0004-637X/749/1/41

    DO - 10.1088/0004-637X/749/1/41

    M3 - Article

    VL - 749

    JO - Astrophysical Journal

    JF - Astrophysical Journal

    SN - 0004-637X

    IS - 1

    M1 - 41

    ER -