loggish: Log-type transformation of a vector

In COINr: Composite Indicator Construction and Analysis

Description

This applies various simple transformations, to be used by the treat() function. This function is probably not very useful on its own because it requires params, a list of parameters which are used to output the type and status of treatment applied.

Usage

loggish(x, ltype, params = NULL)

Arguments

x	A vector or column of data to transform
ltype	The type of log transformation - see deflog in treat().
params	Some extra parameters to pass. These parameters mostly concern internal messages for treat() and this can be left unspecified unless ltype == "boxcox", in which case there should be a parameter params$boxlam specified (see BoxCox()). However, if you wish to use a Box Cox transformation, it is better to use BoxCox() directly.

Value

A list with

• .$x is the transformed vector of data

• .$Flag is a flag of the type of treatment specified (used inside treat())

• .$Treatment the treatment applied (used inside treat())

• .$TreatSpec the treatment specified (used inside treat())

See Also

• treat() Outlier treatment

Examples

# get a column of data with outliers
x <- ASEMIndData$Tariff
# apply a GII transformation
xdash <- loggish(x, ltype = "GIIlog")
# plot one against the other
plot(x, xdash$x)

COINr documentation built on Nov. 30, 2021, 9:06 a.m.