01 Frequentist and Bayesian Approaches to Spectral Line Detection in Astronomy

Oregon Convention Center 

The problem of detecting spectral lines is ubiquitous in several areas of astronomy, as they enable measurements of fundamental physical properties of astronomical objects such as distances, chemical compositions, and temperature. In the near future, the Euclid satellite will provide spectra from millions of galaxies for which distances need to be measured from emission lines (a specific kind of spectral line). Current approaches to detect spectral lines in the astronomy literature often lack the ability to control for the inflation of the probability of false discovery due to searching over multiple regions of the spectrum – a statistical phenomenon also referred to in high-energy physics as "the look-elsewhere effect". This project applies existing (frequentist) solutions that handle the look-elsewhere effect to the problem of detecting and identifying a single spectral line. Moreover, we propose a new Bayesian solution to address this problem, and provide a comparison of both procedures through simulation studies to assess their validity and usefulness.

Look-elsewhere effect

Bayesian methods

Astrostatistics 

Astrostatistics Interest Group