Regression for Astronomical Data with Realistic Distributions, Errors and Non-linearity
Preprint
Abstract
We have developed a new regression technique, the maximum likelihood (ML)-based method and its variant, the KS-test based method, designed to obtain unbiased regression results from typical astronomical data. A normalizing flow model is employed to automatically estimate the unobservable intrinsic distribution of the independent variable as well as the unobservable correlation between uncertainty level and intrinsic value of both independent and dependent variables from the observed data points in a variational inference based empirical Bayes approach. By incorporating these estimated distributions, our method comprehensively accounts for the uncertainties associated with both independent and dependent variables. Our test on both mock data and real astronomical data from PHANGS-ALMA and PHANGS-JWST demonstrates that both the ML based method and the KS-test based method significantly outperform the existing widely-used methods, particularly in cases of low signal-to-noise ratios. The KS-test based method exhibits remarkable robustness against deviations from underlying assumptions, complex intrinsic distributions, varying correlations between uncertainty levels and intrinsic values, inaccuracies in uncertainty estimations, outliers, and saturation effects. We recommend the KS-test based method as the preferred choice for general applications, while the ML based method is suggested for small samples with sizes of $N < 100$. A GPU-compatible Python implementation of our methods, nicknamed ``raddest'', will be made publicly available upon acceptance of this paper.
Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Probabilistic Forward Modeling of Galaxy Catalogs with Normalizing FlowsThe Astronomical Journal, 2024
- Marginalised Normal Regression: Unbiased curve fitting in the presence of x-errorsThe Open Journal of Astrophysics, 2023
- The ALMaQUEST Survey – V. The non-universality of kpc-scale star formation relations and the factors that drive themMonthly Notices of the Royal Astronomical Society, 2020
- LEO-Py: Estimating likelihoods for correlated, censored, and uncertain data with given marginal distributionsAstronomy and Computing, 2019
- Hyper-Fit: Fitting Linear Models to Multidimensional Data with Multivariate Gaussian UncertaintiesPublications of the Astronomical Society of Australia, 2015
- The ATLAS3D project – XV. Benchmark for early-type galaxies scaling relations from 260 dynamical models: mass-to-light ratio, dark matter, Fundamental Plane and Mass PlaneMonthly Notices of the Royal Astronomical Society, 2013
- Some Aspects of Measurement Error in Linear Regression of Astronomical DataThe Astrophysical Journal, 2007
- Linear Regression for Astronomical Data with Measurement Errors and Intrinsic ScatterThe Astrophysical Journal, 1996
- Linear regression in astronomy.The Astrophysical Journal, 1990
- Two-dimensional goodness-of-fit testing in astronomyMonthly Notices of the Royal Astronomical Society, 1983