correlation {RGeostats}R Documentation

Plot the [shifted] correlation

Description

Plot the [shifted] correlation, for one or two variables

Usage

correlation(db, name1=1, name2=NA, db2=NA, name.size=NA,
            ndisc=100,flag.plot=TRUE, flag.same=TRUE, flag.regr=FALSE,
            flag.iso=FALSE,flag.diag=FALSE,flag.ce=FALSE,flag.aspoint=FALSE,
            nbr.ce=10,dmin=NA,dmax=NA,dirvect = NA,tolang = NA,opt.code= 0,
            tolcode = 0,mini1=NA, maxi1=NA, mini2=NA, maxi2=NA, size0=0.01,
            reg.lty=1, reg.lwd=1, reg.col= "black", reg.type="l", reg.pch=19,
            ce.lty=1, ce.lwd=1, ce.col= "black", ce.type="l", ce.pch=19,
            diag.lty=1, diag.lwd=1, diag.col= "black", 
	    xlab=NA, ylab=NA, title=NA, verbose=FALSE,...)

Arguments

db

The db-class containing the data information used to calculate the correlation.

name1

Rank of the column containing the first target variable

name2

Rank of the column containing the second target variable

db2

If 'db2' is specified, the second variable is read from this Db. This option is only available when 'flag.same'=TRUE. This Db2 must have exactly the same organization as 'Db'.

name.size

When the argument 'flag.aspoint' has been switched ON, the samples are represented as circles. If the argument 'name.size' is defined (as a valid attribute of db-class provided as first argument), then the samples are represented with a radius proportional to that variable.

ndisc

Number of discretization steps along each axis for graphic representation

flag.plot

TRUE if the correlation plot is produced.

flag.same

TRUE if the variables must be compared at the same location (standard correlation).

FALSE for the shifted correlation.

flag.regr

When TRUE, represent the regression line of the variable 'name2' as a function of the variable 'name1'. Warning: This option is not available for shifted correlation.

flag.iso

When TRUE, the axes are the same, calculated from the minimum of both variables to the maximum of both variables.

flag.diag

When TRUE, represent the first bissector (which corresponds to the diagonal when flag.iso is TRUE)

flag.ce

When TRUE, represent the experimental conditional expectation

flag.aspoint

When TRUE, the different symbols are represented as circles (of constant or varying radii).

When FALSE, the samples are migrated onto an internal fine grid. Then the density of samples per grid node is displayed using a colored raster representation of this internal grid.

Warning: This option is not available for the shifted correlation.

nbr.ce

Number of intervals (along the definition interval of the explanatory variable) used to calculate the experimental conditional expectation.

dmin

Minimum distance for the calculation of the shifted correlation

dmax

Maximum distance for the calculation of the shifted correlation

dirvect

The directions in which the shifted correlation must be calculated. For more information, refer to get.directions

tolang

Angular tolerance. If no angular tolerance is not defined, the tolerance is set to 90 degrees.

opt.code

Option concerning the sample codes for constituting pairs:

  • 0 : a pair of samples is selected whatever their codes (if active)

  • 1 : a pair is retained if the code is active in the data base and the codes of the two samples are closer than tolcode

  • 2 : a pair is retained only if the codes of the two samples are different

tolcode

Maximum distance between the codes of the two samples.

reg.lty

Type of the line used to represent the linear regression.

reg.lwd

Thickness of the line used to represent the linear regression.

reg.col

Color of the line used to represent the linear regression.

reg.type

Type of the line used to represent the linear regression.

reg.pch

Type of symbol used to represent the linear regression (only if reg.type="b")

diag.lty

Type of the line used to represent the first bissector.

diag.lwd

Thickness of the line used to represent the first bissector.

diag.col

Color of the line used to represent the first bissector.

ce.lty

Type of the line used to represent the linear regression.

ce.lwd

Thickness of the line used to represent the experimental conditional expectation.

ce.col

Color of the line used to represent the experimental conditional expectation.

ce.type

Type of the line used to represent the experimental conditional expectation.

ce.pch

Type of symbol used to represent the experimental conditional expectation. (only if ce.type="b")

mini1

Minimum value for the first target variable

maxi1

Maximum value for the first target variable

mini2

Minimum value for the second target variable

maxi2

Maximum value for the second target variable

size0

Constant size assigned to the symbols (only used if flag.aspoint=TRUE)

xlab,ylab

Labels for the graphic, arguments used by the Splus function "plot"

title

Title to be displayed (optional)

verbose

When TRUE, the calculation of the correlation and (optionally) the regression line produces verbose outputs.

...

Parameters transmitted to the functions db.plot, lines and symbols.

Details

When one variable is provided, the function calculates: Correl(Z1(x),Z1(x+h)) For two variables, the function calculates: Correl(Z1(x),Z2(x+h))

Value

The correlation coefficient.


[Package RGeostats version 11.1.2 Index]