# Statistics

Geospatial statistics can refer to statistics computed with samples collected in a spatial domain (a.k.a. geospatial data), or to statistics computed with multiple realizations of a random field (a.k.a. ensemble).

## Data

The following statistics have geospatial semantics (i.e. make use of spatial coordinates), and as such, approximate better geospatial variables when compared to their non-geospatial counterparts:

Statistics.meanMethod
mean(sdata)
mean(sdata, v)
mean(sdata, v, s)

Spatial mean of spatial data sdata. Optionally, specify the variable v and the block side s.

Statistics.varMethod
var(sdata)
var(sdata, v)
var(sdata, v, s)

Spatial variance of spatial data sdata. Optionally, specify the variable v and the block side s.

Statistics.quantileMethod
quantile(sdata, p)
quantile(sdata, v, p)
quantile(sdata, v, p, s)

Spatial quantile of spatial data sdata at probability p. Optionally, specify the variable v and the block side s.

A histogram with geospatial semantics is also available where the heights of the bins are readjusted based on the coordinates of the samples (i.e. declustered histogram):

GeoStatsBase.EmpiricalHistogramType
EmpiricalHistogram(sdata, var, [s]; kwargs...)

Spatial histogram of variable var in spatial data sdata. Optionally, specify the block side s and the keyword arguments kwargs for the fit(Histogram, ...) call.

## Ensemble

A set of geostatistical realizations of a random field represents a probability distribution. It is often useful to compute summary statistics with this set or ensemble: