Least Squares Two-Point Function Estimation

TL;DR

This paper introduces a novel least squares estimator for the two-point function of homogeneous and isotropic random fields, allowing direct estimation at specific distances rather than binned averages.

Contribution

It proposes a new interpolation-based least squares method for more precise two-point function estimation at specific distances.

Findings

Enables direct estimation at specific distances

Improves over traditional binned averaging methods

Applicable to homogeneous and isotropic random fields

Abstract

The standard estimator for the two-point function of a homogeneous and isotropic random field is a special case of a larger class of least squares estimators that interpolate the function values. Using a different interpolation scheme, two-point function values can be estimated at specific distances, instead of the binned averages.