On the repeated inversion of a covariance matrix

TL;DR

This paper presents an efficient method for removing outliers from data sets while considering correlations between data points, improving the accuracy of likelihood-based parameter estimation.

Contribution

It introduces a novel procedure for outlier elimination that accounts for data correlations, enhancing robustness in likelihood maximization.

Findings

The method effectively identifies outliers in correlated data.

It improves parameter estimation accuracy.

The approach is computationally efficient.

Abstract

In many cases, the values of some model parameters are determined by maximising the likelihood of a set of data points given the parameter values. The presence of outliers in the data and correlations between data points complicate this procedure. An efficient procedure for the elimination of outliers is presented which takes the correlations between data points into account.