arXiv subjects

David W. Hogg

Publications and source records attributed to David W. Hogg.

At least 73 records · Page 4Linked to original sources

arXiv2021

Selection Functions in Astronomical Data Modeling, with the Space Density of White Dwarfs as Worked Example

Statistical studies of astronomical data sets, in particular of cataloged properties for discrete objects, are central to astrophysics. One cannot model those objects' population properties or incidences without a quantitative understanding of the conditions under which these objects ended up in a catalog or sample, the sample's selection function. As systematic and didactic introductions to this topic are scarce in the astrophysical literature, we aim to provide one, addressing generically the following questions: What is a selection function? What arguments $\vec{q}$ should a selection function depend on? Over what domain must a selection function be defined? What approximations and simplifications can be made? And, how is a selection function used in `modelling'? We argue that volume-complete samples, with the volume drastically curtailed by the faintest objects, reflect a highly sub-optimal selection function that needlessly reduces the number of bright and usually rare objects in the sample. We illustrate these points by a worked example, deriving the space density of white dwarfs (WD) in the Galactic neighbourhood as a function of their luminosity and Gaia color, $Φ_0(M_G,B-R)$ in [mag$^{-2}$pc$^{-3}$]. We construct a sample of $10^5$ presumed WDs through straightforward selection cuts on the Gaia EDR3 catalog, in magnitude, color, parallax, and astrometric fidelity $\vec{q}=(m_G,B-R,\varpi,p_{af})$. We then combine a simple model for $Φ_0$ with the effective survey volume derived from this selection function $S_C(\vec{q})$ to derive a detailed and robust estimate of $Φ_0(M_G,B-R)$. This resulting white dwarf luminosity-color function $Φ_0(M_G,B-R)$ differs dramatically from the initial number density distribution in the luminosity-color plane: by orders of magnitude in density and by four magnitudes in density peak location.

astro-ph.IM↗

arXiv2021

Orbital Torus Imaging: Using Element Abundances to Map Orbits and Mass in the Milky Way

Many approaches to galaxy dynamics assume that the gravitational potential is simple and the distribution function is time-invariant. Under these assumptions there are traditional tools for inferring potential parameters given observations of stellar kinematics (e.g., Jeans models). However, spectroscopic surveys measure many stellar properties beyond kinematics. Here we present a new approach for dynamical inference, Orbital Torus Imaging, which makes use of kinematic measurements and element abundances (or other invariant labels). We exploit the fact that, in steady state, stellar labels vary systematically with orbit characteristics (actions), yet must be invariant with respect to orbital phases (conjugate angles). The orbital foliation of phase space must therefore coincide with surfaces along which all moments of all stellar label distributions are constant. Both classical-statistics and Bayesian methods can be built on this; these methods will be more robust and require fewer assumptions than traditional tools because they require no knowledge of the (spatial) survey selection function and they do not involve second moments of velocity distributions. We perform a classical-statistics demonstration with red giant branch stars from the APOGEE surveys: We model the vertical orbit structure in the Milky Way disk to constrain the local disk mass, scale height, and the disk--halo mass ratio (at fixed local circular velocity). We find that the disk mass can be constrained (naïvely) at the few-percent level with Orbital Torus Imaging using only eight element-abundance ratios, demonstrating the promise of combining stellar labels with dynamical invariants.

astro-ph.GA↗

arXiv2021

An unsupervised method for identifying $X$-enriched stars directly from spectra: Li in LAMOST

Stars with peculiar element abundances are important markers of chemical enrichment mechanisms. We present a simple method, tangent space projection (TSP), for the detection of $X$-enriched stars, for arbitrary elements $X$, even from blended lines. Our method does not require stellar labels, but instead directly estimates the counterfactual unrenriched spectrum from other unlabelled spectra. As a case study, we apply this method to the $6708~$Å Li doublet in LAMOST DR5, identifying 8,428 Li-enriched stars seamlessly across evolutionary state. We comment on the explanation for Li-enrichement for different subpopulations, including planet accretion, nonstandard mixing, and youth.

astro-ph.SR↗

arXiv2021

Mapping stellar surfaces I: Degeneracies in the rotational light curve problem

Thanks to missions like Kepler and TESS, we now have access to tens of thousands of high precision, fast cadence, and long baseline stellar photometric observations. In principle, these light curves encode a vast amount of information about stellar variability and, in particular, about the distribution of starspots and other features on their surfaces. Unfortunately, the problem of inferring stellar surface properties from a rotational light curve is famously ill-posed, as it often does not admit a unique solution. Inference about the number, size, contrast, and location of spots can therefore depend very strongly on the assumptions of the model, the regularization scheme, or the prior. The goal of this paper is twofold: (1) to explore the various degeneracies affecting the stellar light curve "inversion" problem and their effect on what can and cannot be learned from a stellar surface given unresolved photometric measurements; and (2) to motivate ensemble analyses of the light curves of many stars at once as a powerful data-driven alternative to common priors adopted in the literature. We further derive novel results on the dependence of the null space on stellar inclination and limb darkening and show that single-band photometric measurements cannot uniquely constrain quantities like the total spot coverage without the use of strong priors. This is the first in a series of papers devoted to the development of novel algorithms and tools for the analysis of stellar light curves and spectral time series, with the explicit goal of enabling statistically robust inference about their surface properties.

astro-ph.SR↗

arXiv2021

Two-point statistics without bins: A continuous-function generalization of the correlation function estimator for large-scale structure

The two-point correlation function (2pcf) is the key statistic in structure formation; it measures the clustering of galaxies or other density field tracers. Estimators of the 2pcf, including the standard Landy-Szalay (LS) estimator, evaluate the 2pcf in hard-edged separation bins, which is scientifically inappropriate and results in a poor trade-off between bias and variance. We present a new 2pcf estimator, the Continuous-Function Estimator, which generalizes LS to a continuous representation and obviates binning in separation or any other pair property. Our estimator, inspired by the mathematics of least-squares fitting, replaces binned pair counts with projections onto basis functions; it outputs the best linear combination of basis functions to describe the 2pcf. The choice of basis can take into account the expected form of the 2pcf, as well as its dependence on pair properties other than separation. We show that the Continuous-Function Estimator with a cubic-spline basis better represents the shape of the 2pcf compared to LS. We also estimate directly the baryon acoustic scale, using a small number of physically-motivated basis functions. Critically, this leads to a reduction in the number of mock catalogs required for covariance estimation, which is currently the limiting step in many 2pcf analyses. We discuss further applications of the Continuous-Function Estimator, including determination of the dependence of clustering on galaxy properties and searches for potential inhomogeneities or anisotropies in large-scale structure.

astro-ph.CO↗

arXiv2021

Fitting very flexible models: Linear regression with large numbers of parameters

There are many uses for linear fitting; the context here is interpolation and denoising of data, as when you have calibration data and you want to fit a smooth, flexible function to those data. Or you want to fit a flexible function to de-trend a time series or normalize a spectrum. In these contexts, investigators often choose a polynomial basis, or a Fourier basis, or wavelets, or something equally general. They also choose an order, or number of basis functions to fit, and (often) some kind of regularization. We discuss how this basis-function fitting is done, with ordinary least squares and extensions thereof. We emphasize that it is often valuable to choose far more parameters than data points, despite folk rules to the contrary: Suitably regularized models with enormous numbers of parameters generalize well and make good predictions for held-out data; over-fitting is not (mainly) a problem of having too many parameters. It is even possible to take the limit of infinite parameters, at which, if the basis and regularization are chosen correctly, the least-squares fit becomes the mean of a Gaussian process. We recommend cross-validation as a good empirical method for model selection (for example, setting the number of parameters and the form of the regularization), and jackknife resampling as a good empirical method for estimating the uncertainties of the predictions made by the model. We also give advice for building stable computational implementations.

physics.data-an↗

arXiv2020

Principled point-source detection in collections of astronomical images

We review the well-known matched filter method for the detection of point sources in astronomical images. This is shown to be optimal (that is, to saturate the Cramer--Rao bound) under stated conditions that are very strong: an isolated source in background-dominated imaging with perfectly known background level, point-spread function, and noise models. We show that the matched filter produces a maximum-likelihood estimate of the brightness of a purported point source, and this leads to a simple way to combine multiple images---taken through the same bandpass filter but with different noise levels and point-spread functions---to produce an optimal point source detection map. We then extend the approach to images taken through different bandpass filters, introducing the SED-matched filter, which allows us to combine images taken through different filters, but requires us to specify the colors of the objects we wish to detect. We show that this approach is superior to some methods traditionally employed, and that other traditional methods can be seen as instances of SED-matched filtering with implied (and often unreasonable) priors. We present a Bayesian formulation, including a flux prior that leads to a closed-form expression with low computational cost.

astro-ph.IM↗

arXiv2020

The Strength of the Dynamical Spiral Perturbation in the Galactic Disk

The mean Galactocentric radial velocities $\langle v_{R}\rangle(R,φ)$ of luminous red giant stars within the mid-plane of the Milky Way reveal a spiral signature, which could plausibly reflect the response to a non-axisymmetric perturbation of the gravitational potential in the Galactic disk. We apply a simple steady-state toy model of a logarithmic spiral to interpret these observations, and find a good qualitative and quantitative match. Presuming that the amplitude of the gravitational potential perturbation is proportionate to that in the disk's surface mass density, we estimate the surface mass density amplitude to be $Σ_{\rm max} (R_{\odot})\approx 5.5\,\rm M_{\odot}\,pc^{-2}$ at the solar radius when choosing a fixed pattern speed of $Ω_{\mathrm p}=12\,\rm km\,s^{-1}\,kpc^{-1}$. Combined with the local disk density, this implies a surface mass density contrast between the arm and inter-arm regions of approximately $\pm 10\%$ at the solar radius, with an increases towards larger radii. Our model constrains the pitch angle of the dynamical spiral arms to be approximately $12^{\circ}$.

astro-ph.GA↗

arXiv2020

How to obtain the redshift distribution from probabilistic redshift estimates

A trustworthy estimate of the redshift distribution $n(z)$ is crucial for using weak gravitational lensing and large-scale structure of galaxy catalogs to study cosmology. Spectroscopic redshifts for the dim and numerous galaxies of next-generation weak-lensing surveys are expected to be unavailable, making photometric redshift (photo-$z$) probability density functions (PDFs) the next-best alternative for comprehensively encapsulating the nontrivial systematics affecting photo-$z$ point estimation. The established stacked estimator of $n(z)$ avoids reducing photo-$z$ PDFs to point estimates but yields a systematically biased estimate of $n(z)$ that worsens with decreasing signal-to-noise, the very regime where photo-$z$ PDFs are most necessary. We introduce Cosmological Hierarchical Inference with Probabilistic Photometric Redshifts (CHIPPR), a statistically rigorous probabilistic graphical model of redshift-dependent photometry, which correctly propagates the redshift uncertainty information beyond the best-fit estimator of $n(z)$ produced by traditional procedures and is provably the only self-consistent way to recover $n(z)$ from photo-$z$ PDFs. We present the $\texttt{chippr}$ prototype code, noting that the mathematically justifiable approach incurs computational expense. The CHIPPR approach is applicable to any one-point statistic of any random variable, provided the prior probability density used to produce the posteriors is explicitly known; if the prior is implicit, as may be the case for popular photo-$z$ techniques, then the resulting posterior PDFs cannot be used for scientific inference. We therefore recommend that the photo-$z$ community focus on developing methodologies that enable the recovery of photo-$z$ likelihoods with support over all redshifts, either directly or via a known prior probability density.

astro-ph.CO↗

arXiv2020

The power of co-ordinate transformations in dynamical interpretations of Galactic structure

$Gaia$ DR2 has provided an unprecedented wealth of information about the positions and motions of stars in our Galaxy, and has highlighted the degree of disequilibria in the disc. As we collect data over a wider area of the disc it becomes increasingly appealing to start analysing stellar actions and angles, which specifically label orbit space, instead of their current phase space location. Conceptually, while $\bar{x}$ and $\bar{v}$ tell us about the potential and local interactions, grouping in action puts together stars that have similar frequencies and hence similar responses to dynamical effects occurring over several orbits. Grouping in actions and angles refines this further to isolate stars which are travelling together through space and hence have shared histories. Mixing these coordinate systems can confuse the interpretation. For example, it has been suggested that by moving stars to their guiding radius, the Milky Way spiral structure is visible as ridge-like overdensities in the $Gaia$ data \citep{Khoperskov+19b}. However, in this work, we show that these features are in fact the known kinematic moving groups, both in the $L_z-ϕ$ and the $v_{\mathrm{R}}-v_ϕ$ planes. Using simulations we show how this distinction will become even more important as we move to a global view of the Milky Way. As an example, we show that the radial velocity wave seen in the Galactic disc in $Gaia$ and APOGEE should become stronger in the action-angle frame, and that it can be reproduced by transient spiral structure.

astro-ph.GA↗

arXiv2020

Data Analysis Recipes: Products of multivariate Gaussians in Bayesian inferences

A product of two Gaussians (or normal distributions) is another Gaussian. That's a valuable and useful fact! Here we use it to derive a refactoring of a common product of multivariate Gaussians: The product of a Gaussian likelihood times a Gaussian prior, where some or all of those parameters enter the likelihood only in the mean and only linearly. That is, a linear, Gaussian, Bayesian model. This product of a likelihood times a prior pdf can be refactored into a product of a marginalized likelihood (or a Bayesian evidence) times a posterior pdf, where (in this case) both of these are also Gaussian. The means and variance tensors of the refactored Gaussians are straightforward to obtain as closed-form expressions; here we deliver these expressions, with discussion. The closed-form expressions can be used to speed up and improve the precision of inferences that contain linear parameters with Gaussian priors. We connect these methods to inferences that arise frequently in physics and astronomy. If all you want is the answer, the question is posed and answered at the beginning of Section 3. We show two toy examples, in the form of worked exercises, in Section 4. The solutions, discussion, and exercises in this Note are aimed at someone who is already familiar with the basic ideas of Bayesian inference and probability.

stat.CO↗

arXiv2020

Temperatures and Metallicities of M dwarfs in the APOGEE Survey

M dwarfs have enormous potential for our understanding of structure and formation on both Galactic and exoplanetary scales through their properties and compositions. However, current atmosphere models have limited ability to reproduce spectral features in stars at the coolest temperatures ($T_{\rm eff} < 4200\,$K) and to fully exploit the information content of current and upcoming large-scale spectroscopic surveys. Here we present a catalog of spectroscopic temperatures, metallicities, and spectral types for 5875 M dwarfs in the Apache Point Observatory Galactic Evolution Experiment (APOGEE) and Gaia DR2 surveys using The Cannon: a flexible, data-driven spectral-modeling and parameter-inference framework demonstrated to estimate stellar-parameter labels ($T_{\rm eff}$, log$g$, [Fe/H], and detailed abundances) to high precision. Using a training sample of 87 M dwarfs with optically derived labels spanning $2860 < T_{\rm eff} < 4130\,$K calibrated with bolometric temperatures, and $-0.5 < $[Fe/H]$ < 0.5\,$dex calibrated with FGK binary metallicities, we train a two-parameter model with predictive accuracy (in cross-validation) to $77\,$K and $0.09\,$dex respectively. We also train a one-dimensional spectral classification model using 51 M dwarfs with Sloan Digital Sky Survey optical spectral types ranging from M0 to M6, to predictive accuracy of 0.7 types. We find Cannon temperatures to be in agreement to within $60\,$K compared to a subsample of 1702 sources with color-derived temperatures, and Cannon metallicities to be in agreement to within $0.08\,$dex metallicity compared to a subsample of 15 FGK+M or M+M binaries. Finally, our comparison between Cannon and APOGEE pipeline (ASPCAP DR14) labels finds that ASPCAP is systematically biased toward reporting higher temperatures and lower metallicities for M dwarfs.

astro-ph.SR↗

arXiv2020

Forward modeling the orbits of companions to pulsating stars from their light travel time variations

Mutual gravitation between a pulsating star and an orbital companion leads to a time-dependent variation in path length for starlight traveling to Earth. These variations can be used for coherently pulsating stars, such as the δ Scuti variables, to constrain the masses and orbits of their companions. Observing these variations for δ Scuti stars has previously relied on subdividing the light curve and measuring the average pulsation phase in equally sized subdivisions, which leads to under-sampling near periapsis. We introduce a new approach that simultaneously forward-models each sample in the light curve and show that this method improves upon current sensitivity limits - especially in the case of highly eccentric and short-period binaries. We find that this approach is sensitive enough to observe Jupiter mass planets around δ Scuti stars under ideal conditions, and use gravity-mode pulsations in the subdwarf B star KIC 7668647 to detect its companion without radial velocity data. We further provide robust detection limits as a function of the SNR of the pulsation mode and determine that the minimum detectable light travel time amplitude for a typical Kepler δ Scuti is around 2 s. This new method significantly enhances the application of light travel time variations to detecting short period binaries with pulsating components, and pulsating A-type exoplanet host stars, especially as a tool for eliminating false positives.

astro-ph.SR↗

arXiv2020

Close Binary Companions to APOGEE DR16 Stars: 20,000 Binary-star Systems Across the Color-Magnitude Diagram

Many problems in contemporary astrophysics---from understanding the formation of black holes to untangling the chemical evolution of galaxies---rely on knowledge about binary stars. This, in turn, depends on discovery and characterization of binary companions for large numbers of different kinds of stars in different chemical and dynamical environments. Current stellar spectroscopic surveys observe hundreds of thousands to millions of stars with (typically) few observational epochs, which allows binary discovery but makes orbital characterization challenging. We use a custom Monte Carlo sampler (The Joker) to perform discovery and characterization of binary systems through radial-velocities, in the regime of sparse, noisy, and poorly sampled multi-epoch data. We use it to generate posterior samplings in Keplerian parameters for 232,531 sources released in APOGEE Data Release 16. Our final catalog contains 19,635 high-confidence close-binary (P < few years, a < few AU) systems that show interesting relationships between binary occurrence rate and location in the color-magnitude diagram. We find notable faint companions at high masses (black-hole candidates), at low masses (substellar candidates), and at very close separations (mass-transfer candidates). We also use the posterior samplings in a (toy) hierarchical inference to measure the long-period binary-star eccentricity distribution. We release the full set of posterior samplings for the entire parent sample of 232,531 stars. This set of samplings involves no heuristic "discovery" threshold and therefore can be used for myriad statistical purposes, including hierarchical inferences about binary-star populations and sub-threshold searches.

astro-ph.SR↗

arXiv2020

High-resolution spectroscopy of the GD-1 stellar stream localizes the perturber near the orbital plane of Sagittarius

The $100^\circ$-long thin stellar stream in the Milky Way halo, GD-1, has an ensemble of features that may be due to dynamical interactions. Using high-resolution MMT/Hectochelle spectroscopy we show that a spur of GD-1-like stars outside of the main stream are kinematically and chemically consistent with the main stream. In the spur, as in the main stream, GD-1 has a low intrinsic radial velocity dispersion, $σ_{V_r}\lesssim1\,\rm km\,s^{-1}$, is metal-poor, $\rm [Fe/H]\approx-2.3$, with little $\rm [Fe/H]$ spread and some variation in $\rm [α/Fe]$ abundances, which point to a common globular cluster progenitor. At a fixed location along the stream, the median radial velocity offset between the spur and the main stream is smaller than $0.5\,\rm km\,s^{-1}$, comparable to the measurement uncertainty. A flyby of a massive, compact object can change orbits of stars in a stellar stream and produce features like the spur observed in GD-1. In this scenario, the radial velocity of the GD-1 spur relative to the stream constrains the orbit of the perturber and its current on-sky position to $\approx5,000\,\rm deg^2$. The family of acceptable perturber orbits overlaps the stellar and dark-matter debris of the Sagittarius dwarf galaxy in present-day position and velocity. This suggests that GD-1 may have been perturbed by a globular cluster or an extremely compact dark-matter subhalo formerly associated with Sagittarius.

astro-ph.GA↗

arXiv2019

emcee v3: A Python ensemble sampling toolkit for affine-invariant MCMC

emcee is a Python library implementing a class of affine-invariant ensemble samplers for Markov chain Monte Carlo (MCMC). This package has been widely applied to probabilistic modeling problems in astrophysics where it was originally published, with some applications in other fields. When it was first released in 2012, the interface implemented in emcee was fundamentally different from the MCMC libraries that were popular at the time, such as PyMC, because it was specifically designed to work with "black box" models instead of structured graphical models. This has been a popular interface for applications in astrophysics because it is often non-trivial to implement realistic physics within the modeling frameworks required by other libraries. Since emcee's release, other libraries have been developed with similar interfaces, such as dynesty (Speagle 2019). The version 3.0 release of emcee is the first major release of the library in about 6 years and it includes a full re-write of the computational backend, several commonly requested features, and a set of new "move" implementations.

astro-ph.IM↗

arXiv2019

Wobble: A Data-driven Analysis Technique for Time-series Stellar Spectra

In recent years, dedicated extreme-precision radial velocity (EPRV) spectrographs have produced vast quantities of high-resolution, high-signal-to-noise time-series spectra for bright stars. These data contain valuable information for the dual purposes of planet detection via the measured RVs and stellar characterization via the co-added spectra. However, considerable data analysis challenges exist in extracting these data products from the observed spectra at the highest possible precision, including the issue of poorly-characterized telluric absorption features and the common use of an assumed stellar spectral template. In both of these examples, precision-limiting reliance on external information can be sidestepped using the data directly. Here we propose a data-driven method to simultaneously extract precise RVs and infer the underlying stellar and telluric spectra using a linear model (in the log of flux). The model employs a convex objective and convex regularization to keep the optimization of the spectral components fast. We implement this method in wobble, an open-source python package which uses TensorFlow in one of its first non-neural-network applications to astronomical data. In this work, we demonstrate the performance of wobble on archival HARPS spectra. We recover the canonical exoplanet 51 Pegasi b, detect the secular RV evolution of the M dwarf Barnard's Star, and retrieve the Rossiter-McLaughlin effect for the Hot Jupiter HD 189733b. The method additionally produces extremely high-S/N composite stellar spectra and detailed time-variable telluric spectra, which we also present here.

astro-ph.IM↗

arXiv2019

The K2 Bright Star Survey I: Methodology and Data Release

While the Kepler Mission was designed to look at tens of thousands of faint stars (V > 12), brighter stars that saturated the detector are important because they can be and have been observed very accurately by other instruments. By analyzing the unsaturated scattered-light `halo' around these stars, we have retrieved precise light curves of most of the brightest stars in K2 fields from Campaign~4 onwards. The halo method does not depend on the detailed cause and form of systematics, and we show that it is effective at extracting light curves from both normal and saturated stars. The key methodology is to optimize the weights of a linear combination of pixel time series with respect to an objective function. We test a range of such objective functions, finding that lagged Total Variation, a generalization of Total Variation, performs well on both saturated and unsaturated K2 targets. Applying this to the bright stars across the K2 Campaigns reveals stellar variability ubiquitously, including effects of stellar pulsation, rotation, and binarity. We describe our pipeline and present a catalogue of the 161 bright stars, with classifications of their variability, asteroseismic parameters for red giants with well-measured solar-like oscillations, and remarks on interesting objects. These light curves are publicly available as a High Level Science Product from the Mikulski Archive for Space Telescopes (MAST).

astro-ph.SR↗