Use of the Jackknife Method to Estimate Autocorrelation Functions (or Variograms)

Abstract

An application of the jackknife method to estimate autocorrelation functions (or variograms) of stationary random functions is discussed. Using the jackknife estimators and the corresponding jackknife variances, the models of the autocorrelation functions are fitted by the weighted least squares method. The method is particularly effective to study robustness of the estimators when the number of data points is small. The technique is applied to simulated data sets with known autocorrelation functions.