E. Pardo-Iguzquiza P. A. Dowd Ricardo A. Olea 2015 <p>The bootstrap is a computer-intensive resampling method for estimating<br>the uncertainty of complex statistical models. We expand on an<br>application of the bootstrap for inferring semivariogram parameters and<br>their uncertainty. The model fitted to the median of the bootstrap distribution<br>of the experimental semivariogram is proposed as an estimator of<br>the semivariogram. The proposed application is not restricted to normal<br>data and the estimator is resistant to outliers. Improvements are more<br>significant for data-sets with less than 100 observations, which are<br>those for which semivariogram model inference is the most difficult. The<br>application is illustrated by using it to characterize a synthetic random<br>field for which the true semivariogram type and parameters are known.</p> application/pdf en Southern African Institute of Mining and Metallurgy (SAIMM) Robust and resistant semivariogram modelling using a generalized bootstrap article