Polar polarization: a new method for polarimetry analysis

TL;DR

This paper introduces a new likelihood-based analysis method for polarization transfer measurements that reduces uncertainties, improves data representation, and enhances comparison with theoretical models.

Contribution

A novel likelihood maximization approach for polarization analysis that accounts for multi-parameter models and provides smooth data representation.

Findings

Reduced statistical and systematic uncertainties in simulated data

Enhanced ability to reveal physical information

Significantly decreased CPU time

Abstract

We present a novel analysis method for measurements of polarization transferred in $A(\vec{e},e'\vec{N})$ experiments, which can be applied to other kinds of polarization measurements as well. In this method the polarization transfer components are presented in spherical coordinates using an efficient likelihood numerical maximization based on an analytic derivation. We also propose a formalism that accounts for multi-parameter models, and which yields a smooth and continuous representation of the data (rather than using standard binning). Applying this method on simulated data generates results with reduced statistical and systematic uncertainties and enables revealing physical information that is lost in standard binning of the data. The obtained results can be compared easily to theoretical models and other measurements. Furthermore, CPU time is significantly reduced using this method.