Fast Bayesian Semiparametric Curve-Fitting and Clustering in Massive Data With Application to Cosmology

TL;DR

This paper introduces a Bayesian semiparametric mixture model for efficient curve-fitting and clustering in massive cosmological datasets, revealing key features like change points and quasar clustering at high redshift.

Contribution

It presents a novel Bayesian mixture model approach for curve-fitting and clustering in large-scale cosmological data, with efficient algorithms and new posterior analysis methods.

Findings

Identified four change points in the regression curve.

Detected clustering of quasars at high redshift.

Provided insights into cosmological parameters related to universe acceleration.

Abstract

Recent technological advances have led to a flood of new data on cosmology rich in information about the formation and evolution of the universe, e.g., the data collected in Sloan Digital Sky Survey (SDSS) for more than 200 million objects. The analyses of such data demand cutting edge statistical technologies. Here, we have used the concept of mixture model within Bayesian semiparametric methodology to fit the regression curve with the bivariate data for the apparent magnitude and redshift for Quasars in SDSS (2007) catalogue. Associated with the mixture modeling is a highly efficient curve-fitting procedure, which is central to the application considered in this paper. Moreover, we adopt a new method for analysing the posterior distribution of clusterings, also generated as a by-product of our methodology. The results of our analysis of the cosmological data clearly indicate the existence of four change points on the regression curve andposssibiltiy of clustering of quasars specially at high redshift. This sheds new light not only on the issue of evolution, existence of acceleration or decceleration and environment around quasars at high redshift but also help us to estimate thecosmological parameters related to acceleration or decceleration.