## Local polinomial estimation

Nonparametric estimation of a multidimensional regression function (e.g. a spatial trend), a probability density function or a semivariogram. To speed up computations, linear binning is used to discretize the (corresponding) data. A full bandwidth matrix and a multiplicative triweight kernel is used to compute the weights. Main calculations are performed in FORTRAN using the LAPACK library.

npsp-package

npsp: Nonparametric spatial (geo)statistics

np.fitgeo()

Fit a nonparametric geostatistical model

locpol() locpolhcv()

Local polynomial estimation

np.den()

Local polynomial density estimation

np.svar() np.svariso() np.svariso.hcv() np.svariso.corr()

Local polynomial estimation of the semivariogram

interp() predict(<locpol.bin>) predict(<np.den>)

Fast linear interpolation of a regular grid

np.geo()

Nonparametric geostatistical model (S3 class "np.geo")

## Bandwidth selection

Cross-validation methods for bandwidth selection in local polynomial kernel smoothing.

h.cv() hcv.data()

Cross-validation methods for bandwidth selection

## Linear binning

Methods for multidimensional linear binning.

binning() as.bin.data()

Linear binning

bin.den() as.bin.den()

Linear binning for density estimation

svar.bin() svariso()

Linear binning of semivariances

## Variogram fitting

Flexible (isotropic) Shapiro-Botha variogram model fitting by WLS.

fitsvar.sb.iso()

Fit an isotropic Shapiro-Botha variogram model

disc.sb()

Discretization nodes of a Shapiro-Botha variogram model

## Variogram and covariogram utilities

Variogram models and methods.

sv()

Evaluate a semivariogram model

covar()

Covariance values

varcov()

Covariance matrix

svarmod() svarmod.sb.iso() svarmodels()

Define a (semi)variogram model

kappasb()

Coefficients of an extended Shapiro-Botha variogram model

svar.grid()

Discretize a (semi)variogram model

## Kriging

Nonparametric residual kriging (sometimes called external drift kriging) and simple kriging.

np.kriging() kriging.simple() kriging.simple.solve()

Nonparametric (residual) kriging

## Gridded data

S3 class data.grid and methods.

data.grid()

Gridded data (S3 class "data.grid")

grid.par()

Grid parameters (S3 class "grid.par")

coords()

(spatial) coordinates

coordvalues()

Coordinate values

as.data.grid() as.data.frame(<data.grid>)

data.grid-class methods

interp() predict(<locpol.bin>) predict(<np.den>)

Fast linear interpolation of a regular grid

## Plot

Utilities for plotting data with a continuous color scale and other plot methods.

scattersplot()

Exploratory scatter plots

splot() scolor() jet.colors() hot.colors()

Utilities for plotting with a color scale

spoints()

Scatter plot with a color scale

simage() plot(<np.den>)

Image plot with a color scale

spersp()

Perspective plot with a color scale

plot(<fitgeo>)

Plot a nonparametric geostatistical model

plot(<fitsvar>) plot(<svar.bin>) plot(<np.svar>)

Plot a semivariogram object

image(<data.grid>) persp(<data.grid>) contour(<data.grid>)

R Graphics for gridded data

## Data sets

Data sets included in the package.

aquifer

Wolfcamp aquifer data

earthquakes

Earthquake data

precipitation

Precipitation data

## Other packages

Utilities to interact with other packages.

as.sp()

Convert npsp object to sp object

as.variogram() as.variomodel()

Interface to package "geoR"

as.vgm() vgm.tab.svarmod()

Interface to package "gstat"

## Utilities

Other utilities…

mask()

npsp.tolerance()
rule() rule.binning() rule.svar()
cpu.time()