Title
Calculating Bayesian Confidence Intervals with Stokes Parameters for use in Hard X-Ray Polarimetry
Date of Award
Spring 4-3-2014
Author's School
College of Arts & Sciences
Author's Department/Program
Physics
Abstract
In this paper, we for the first time use Stokes parameters to simulate and analyze measurements in the hard x-ray band. Simulations generate photons according to the Klein-Nishina cross section, and are analyzed by summing the Stokes parameters of individual photons to reconstruct the polarization fraction and angle of the signal. We also offer a consideration of a more common method of polarization analysis, that of fitting a sine wave to the distribution of measured azimuthal angles. This fitting method introduces systematic errors dependent on binning and fitting in the azimuthal angle distribution which can be avoided by using the Stokes parameters. In addition to introducing an analysis of the x-ray domain with Stokes parameters, we present a method of Bayesian inversion to generate a posteriori probability distributions given a distribution of measured values. Such a technique addresses the problems of measuring most-probable and mean polarization values greater than the true value at low polarization.
Language
English (en)
Advisor/Committee Chair
Henric Krawczynski
Advisor/Committee Chair's Department
Department of Physics
Second Advisor
Fabian Kislat
Second Advisor's Department
Department of Physics
Recommended Citation
Clark, Brian A., "Calculating Bayesian Confidence Intervals with Stokes Parameters for use in Hard X-Ray Polarimetry" (2014). Undergraduate Theses—Restricted. 38.
https://openscholarship.wustl.edu/undergrad_honors/38