Statistics for Spatial Data
Statistics for Spatial Data is concerned with analyzing spatial data through statistical models. The book unifies many diverse areas by using consistent notation, and delineates clearly the three strongest growth areas - geostatistical data, lattice data and point patterns. It corrects mistakes which have not previously been subject to close scrutiny from statisticians, and gives new results at the frontiers of the subject.
85 pages matching method in this book
Results 1-3 of 85
What people are saying - Write a review
We haven't found any reviews in the usual places.
Spatial Prediction and Kriging
9 other sections not shown
analysis approach approximately assumed assumption bias Chapter coal-ash data cokriging column computed conditional covariance function covariogram Cressie cross-validation decomposition defined denotes distribution ergodic error process example Figure fitted Gaussian process geostatistical given in Section grid intrinsically stationary isotropic Journel kriging predictor kriging variance large-scale variation lattice linear predictor Markov random field Matheron matrix maximum likelihood mean-squared prediction error median median-polish median-polish kriging method minimizes neighborhood nugget effect o.l.s. estimator observations obtained optimal predictor ordinary kriging outliers parameters plot prediction interval problem random process random variables residuals robust estimator scale second-order stationary Section 2.3 semivariogram shows simulation small-scale variation spatial correlation spatial data spatial data analysis spatial dependence spatial locations spatial model spatial prediction spectral spline square stationary process statistical stochastic stochastic process Suppose tion treatment effects trend unbiased universal kriging valid values variogram estimator variogram model vector weights yields zero zero-mean