The Method of Normalized Correlations

Summary: [This abstract is based on the authors' abstract.] The method of normalized correlations (MoNC) is used on the inference of covariance parameters from spatial or temporal data. The method is compared with maximum likelihood estimation and weighted least squares on both simulated and real data. MoNC was shown to be significantly faster and insensitive to sample size for regularly spaced data, making it potentially useful for calculations involving large data sets and fast preconditioning.