Marginalised Normal Regression: Unbiased curve fitting in the presence of x-errors

Abstract

The history of the seemingly simple problem of straight line fitting in the presence of both $x$ and $y$ errors has been fraught with misadventure, with statistically ad hoc and poorly tested methods abounding in the literature. The problem stems from the emergence of latent variables describing the "true" values of the independent variables, the priors on which have a significant impact on the regression result. By analytic calculation of maximum a posteriori values and biases, and comprehensive numerical mock tests, we assess the quality of possible priors. In the presence of intrinsic scatter, the only prior that we find to give reliably unbiased results in general is a mixture of one or more Gaussians with means and variances determined as part of the inference. We find that a single Gaussian is typically sufficient and dub this model Marginalised Normal Regression (MNR). We illustrate the necessity for MNR by comparing it to alternative methods on an important linear relation in cosmology, and extend it to nonlinear regression and an arbitrary covariance matrix linking $x$ and $y$. We publicly release a Python/Jax implementation of MNR and its Gaussian mixture model extension that is coupled to Hamiltonian Monte Carlo for efficient sampling, which we call ROXY (Regression and Optimisation with X and Y errors).

arXiv comment14+6 pages, 9 figures; minor revision to match Open J. of Astrophysics accepted version

References in corpus (16)

• SciPy 1.0--Fundamental Algorithms for Scientific Computing in Python
• The Sloan Digital Sky Survey: Technical Summary
• The slope of the black-hole mass versus velocity dispersion correlation
• Planck 2013 results. I. Overview of products and scientific results
• Some Aspects of Measurement Error in Linear Regression of Astronomical Data
• Non-linear Least Squares Fitting in IDL with MPFIT
• Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro
• The Tully-Fisher relations of early-type spiral and S0 galaxies
• On the Correlations of Massive Black Holes with their Host Galaxies
• Fits, and especially linear fits, with errors on both axes, extra variance of the data points and other complications
• Bayesian model selection without evidences: application to the dark energy equation-of-state
• Planck Intermediate Results. XXXVI. Optical identification and redshifts of Planck SZ sources with telescopes in the Canary Islands Observatories
• A Gibbs Sampler for Multivariate Linear Regression
• Optical validation and characterization of Planck PSZ2 sources at the Canary Islands observatories. II. Second year of LP15 observations
• Velocity dispersion and dynamical masses for 388 Galaxy Clusters and groups. Calibrating the $M_{\mathrm{SZ}}$--$M_{\mathrm{dyn}}$ scaling relation for the PSZ2 sample
• Optical validation and characterization of \Planck\ PSZ2 sources at the Canary Islands observatories. I. First year of LP15 observations

Cited by in corpus (9)

• The Inefficiency of Genetic Programming for Symbolic Regression
• Testing the local supervoid solution to the Hubble tension with direct distance tracers
• The Velocity Field Olympics: Assessing velocity field reconstructions with direct distance tracers
• MUSE-DARK III: The evolution of the radial acceleration relation at intermediate redshifts
• No evidence for local $H_0$ anisotropy from Tully--Fisher or supernova distances
• An approach to robust Bayesian regression in astronomy
• A 15 Mpc rotating galaxy filament at redshift z = 0.032
• Regression for Astronomical Data with Realistic Distributions, Errors and Non-linearity
• Forward-modelling Milky Way Cepheids: selection effects and physical priors in the Gaia-HST calibration