Correlation and (weighted) covariance

Compute correlation and (weighted) covariance for multi-layer Raster objects. Like cellStats this function returns a few values, not a Raster* object (see Summary-methods for that).

layerStats(x, stat, w, asSample=TRUE, na.rm=FALSE, ...)

Arguments

x

RasterStack or RasterBrick for which to compute a statistic

stat

Character. The statistic to compute: either 'cov' (covariance), 'weighted.cov' (weighted covariance), or 'pearson' (correlation coefficient)

w

RasterLayer with the weights (should have the same extent, resolution and number of layers as x) to compute the weighted covariance

asSample

Logical. If TRUE, the statistic for a sample (denominator is n-1) is computed, rather than for the population (denominator is n)

na.rm

Logical. Should missing values be removed?

...

Additional arguments (none implemetned)

Value

List with two items: the correlation or (weighted) covariance matrix, and the (weighted) means.

Author

Jonathan A. Greenberg & Robert Hijmans. Weighted covariance based on code by Mort Canty

References

For the weighted covariance:

• Canty, M.J. and A.A. Nielsen, 2008. Automatic radiometric normalization of multitemporal satellite imagery with the iteratively re-weighted MAD transformation. Remote Sensing of Environment 112:1025-1036.

• Nielsen, A.A., 2007. The regularized iteratively reweighted MAD method for change detection in multi- and hyperspectral data. IEEE Transactions on Image Processing 16(2):463-478.

See also

cellStats, cov.wt, weighted.mean

Examples

b <- brick(system.file("external/rlogo.grd", package="raster"))
layerStats(b, 'pearson')
#> $`pearson correlation coefficient`
#>             red     green      blue
#> red   1.0000000 0.9980961 0.9501633
#> green 0.9980961 1.0000000 0.9658011
#> blue  0.9501633 0.9658011 1.0000000
#> 
#> $mean
#>      red    green     blue 
#> 182.2855 185.3509 192.8046 
#> 

layerStats(b, 'cov')
#> $covariance
#>            red    green     blue
#> red   5564.371 5443.405 4993.165
#> green 5443.405 5345.403 4974.478
#> blue  4993.165 4974.478 4962.942
#> 
#> $mean
#>      red    green     blue 
#> 182.2855 185.3509 192.8046 
#> 

# weigh by column number
w <- init(b, v='col')
layerStats(b, 'weighted.cov', w=w)
#> $`weigthed covariance`
#>            red    green     blue
#> red   5670.750 5536.351 5009.851
#> green 5536.351 5427.161 4987.092
#> blue  5009.851 4987.092 4937.007
#> 
#> $`weighted mean`
#>  layer.1  layer.2  layer.3 
#> 177.5983 181.3521 191.5236 
#>