Kriging and Spatial Estimation
Kriging is least-squares estimation built on a variogram. Derive simple kriging from first principles, add the unbiasedness constraint for ordinary kriging, and bring in secondary data with cokriging and external drift, solving the systems live at every step.
20 interactive sections across 1 book
Geostatistics
- Simple kriging from first principles
- Ordinary kriging and the unbiasedness constraint
- Universal kriging and kriging with external drift
- Cokriging with secondary data
- Covariance, correlogram, and variogram, three views of the same thing
- Nugget effect and short-scale variability
- The kriging variance, what it means and what it doesn't
- What makes data spatial?
- From global statistics to local: blocks and panels
- What declustering changes downstream
- h-scatterplots and lag binning
- Indicator variograms
- Robust variogram estimators
- Anisotropic ellipsoids and the search ellipse
- Fitting strategies: by eye, by WLS, by likelihood
- Permissible model families (spherical, exponential, Gaussian)
- Common kriging pathologies and how to spot them
- Neighbourhood selection and search ellipsoids
- Sequential Gaussian Simulation (SGS)
- Indicator variograms revisited