bias {nnDiag}R Documentation

Bias Assessment

Description

Assesses bias by the relationship between observations and predictions, whether in ordered groups or by individual data elements.

Usage

bias(object, mode = "groups")

Arguments

object object of class "nnDgrps"
mode can either be set to "groups" to assess means of groups of observations against means of predictions, or "points" to assess each observation against its corresponding prediction.

Value

An object of class "nnDbias".

When mode = "groups", object is a list that contains the following components:
mean.predictions a vector of means of the groups of predictions.
mean.observations a vector of means of the groups of observations.

When mode = "points" the list contains:
predictions vector of predictions.
observations vector of observations.

Author(s)

Brian Walters walte137@msu.edu

References

McRoberts, R.E. (2009) Diagnostic tools for nearest neighbors techniques when used with satellite imagery, Remote Sensing of Environment. 113, 489–499.

See Also

grouper, arealBias

Examples

data(LuceVolume)
data(LuceVolume_indx)

##First use grouper to make the this-is-escaped-code{ object
grps <- grouper(LuceVolume$ref.volume, LuceVolume$pred.vol_k18, LuceVolume_indx)

##Using mode "groups"
bg <- bias(grps, mode = "groups")
plot(bg)

##Using mode "points"
bp <- bias(grps, mode = "points")
plot(bp)

[Package nnDiag version 0.0-5 Index]