Challenges in Astrostatistics: A High-Energy Astrophysics Perspective

• Activity Number: 47
• Type: Invited
• Date/Time: Sunday, August 3, 2014 : 4:00 PM to 5:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Aneta Siemiginowska*+ and Vinay Kashyap and David Van Dyk and Xiao-Li Meng
• Companies: Harvard/Smithsonian Center for Astrophysics and Harvard-Smithsonian Center for Astrophysics and Imperial College London and Harvard
• Keywords: Bayesian ; astrostatistics ; X-rays ; Poisson

Abstract

Astrostatistics is becoming well-established in High Energy Astrophysics as data collected by modern missions, e.g. the Chandra X-ray Observatory or the Fermi Gamma-Ray Observatory, become difficult to handle with standard methods. The X-ray and gamma-ray data are Poisson in nature and require sophisticated methods to fully explore information contained in the observations. I will describe the specific challenges that have been addressed by our group (CHASC http://hea-www.harvard.edu/AstroStat/ ). These analysis laid the foundation for extensive application of Bayesian methodologies in high energy astrophysics. I will also present some recent examples of Bayesian methods developed and applied by other groups to obtain constraints on the physics of astronomical sources and on cosmology. However, the richness of the data in spectral, spatial and time domains provide continuous analysis challenges. Standard methods treat these domains independently and only a few attempts of combining them have been made to date. I will discuss some issues related to the simultaneous approach which remains the most demanding open challenge, especially in X-ray and gamma-ray analysis.