A data-driven model for spectra: Finding double redshifts in the Sloan Digital Sky Survey
arXiv:1201.3370 · doi:10.1088/0004-637X/753/2/122
Abstract
We present a data-driven method - heteroscedastic matrix factorization, a kind of probabilistic factor analysis - for modeling or performing dimensionality reduction on observed spectra or other high-dimensional data with known but non-uniform observational uncertainties. The method uses an iterative inverse-variance-weighted least-squares minimization procedure to generate a best set of basis functions. The method is similar to principal components analysis, but with the substantial advantage that it uses measurement uncertainties in a responsible way and accounts naturally for poorly measured and missing data; it models the variance in the noise-deconvolved data space. A regularization can be applied, in the form of a smoothness prior (inspired by Gaussian processes) or a non-negative constraint, without making the method prohibitively slow. Because the method optimizes a justified scalar (related to the likelihood), the basis provides a better fit to the data in a probabilistic sense than any PCA basis. We test the method on SDSS spectra, concentrating on spectra known to contain two redshift components: These are spectra of gravitational lens candidates and massive black-hole binaries. We apply a hypothesis test to compare one-redshift and two-redshift models for these spectra, utilizing the data-driven model trained on a random subset of all SDSS spectra. This test confirms 129 of the 131 lens candidates in our sample and all of the known binary candidates, and turns up very few false positives.
arXiv commentAccepted for publication in ApJ
References in corpus (30)
- The Seventh Data Release of the Sloan Digital Sky Survey
- K-corrections and filter transformations in the ultraviolet, optical, and near infrared
- Spectroscopic Target Selection for the Sloan Digital Sky Survey: The Luminous Red Galaxy Sample
- The Sloan Lens ACS Survey. I. A Large Spectroscopically Selected Sample of Massive Early-Type Lens Galaxies
- The Sloan Lens ACS Survey. V. The Full ACS Strong-Lens Sample
- Spectral Classification of Quasars in the Sloan Digital Sky Survey: Eigenspectra; Redshift and Luminosity Effects
- A recoiling supermassive black hole in the quasar SDSSJ092712.65+294344.0?
- A strong redshift dependence of the broad absorption line quasar fraction
- A Candidate Sub-Parsec Supermassive Binary Black Hole System
- Spectral Classification of Galaxies: An Orthogonal Approach
- Distributions of Galaxy Spectral Types in the Sloan Digital Sky Survey
- Bursty stellar populations and obscured AGN in galaxy bulges
- A systematic search for massive black hole binaries in SDSS spectroscopic sample
- SDSS J092712.65+294344.0: Recoiling Black Hole or A Sub-parsec Binary Candidate?
- Average spectra of massive galaxies in the SDSS
- Creating Spectral Templates from Multicolor Redshift Surveys
- Spectral Templates from Multicolor Redshift Surveys
- SDSSJ092712.65+294344.0: a candidate massive black hole binary
- Automatic Redshift Determination by use of Principal Component Analysis --- I: Fundamentals
- A Robust Classification of Galaxy Spectra: Dealing with Noisy and Incomplete Data
- The Quasar SDSS J105041.35+345631.3: Black Hole Recoil or Extreme Double-Peaked Emitter?
- Reducing the Dimensionality of Data: Locally Linear Embedding of Sloan Galaxy Spectra
- Constraints on the star formation histories of galaxies from z~1 to z~0
- A Principal Component Analysis approach to the Star Formation History of elliptical galaxies in Compact Groups
- The peculiar optical spectrum of 4C+22.25: Imprint of a massive black hole binary?
- Reliable Eigenspectra for New Generation Surveys
- Accurate parameter estimation for star formation history in galaxies using SDSS spectra
- Exploring the Spectral Space of Low Redshift QSOs
- Unusual Double-peaked Emission in the SDSS Quasar J093201.60+031858.7
- A data-driven Bayesian approach for finding young stellar populations in early-type galaxies from their UV-optical spectra
Cited by in corpus (22)
- Spectral Classification and Redshift Measurement for the SDSS-III Baryon Oscillation Spectroscopic Survey
- The Optical-Infrared Extinction Curve and its Variation in the Milky Way
- A systematic search for massive black hole binaries in SDSS spectroscopic sample
- The weirdest SDSS galaxies: results from an outlier detection algorithm
- Principal Component Analysis with Noisy and/or Missing Data
- Weighted principal component analysis: a weighted covariance eigendecomposition approach
- The nature of massive black hole binary candidates: I. Spectral properties and evolution
- Search of sub-parsec massive binary black holes through line diagnosis
- Search of sub-parsec massive binary black holes through line diagnosis II
- Archetype-Based Redshift Estimation for the Dark Energy Spectroscopic Instrument Survey
- S4: A Spatial-Spectral model for Speckle Suppression
- Excalibur: A Non-Parametric, Hierarchical Wavelength-Calibration Method for a Precision Spectrograph
- Joint Bayesian Estimation of Quasar Continua and the Lyman-Alpha Forest Flux Probability Distribution Function
- The Dust Extinction Curve: Beyond R(V)
- Determination of astrophysical parameters of quasars within the Gaia mission
- Galaxy Spectra Networks (GaSNet). III. Generative pre-trained network for spectrum reconstruction, redshift estimate and anomaly detection
- Spectral components analysis of diffuse emission processes
- A Search for Spectral Galaxy Pairs of Overlapping Galaxies based on Fuzzy Recognition
- Redshift determination through weighted phase correlation: a linearithmic implementation
- Reconstructing redshift distributions with photometric galaxy clustering
- ALPCAH: Sample-wise Heteroscedastic PCA with Tail Singular Value Regularization
- Accurate spectroscopic redshift estimation using non-negative matrix factorization: application to MUSE spectra