Gower.dist {shipunov}

Gower distance

Description

Calculates Gower distance

Usage

Gower.dist(data.x, data.y=data.x, rngs=NULL,
 KR.corr=TRUE, na.rm=FALSE)

Arguments

data.x

A matrix or a data frame containing variables that should be used in the computation of the distance.

data.y

A numeric matrix or data frame with the same variables, of the same type, as those in 'data.x'

rngs

A vector with the ranges to scale the variables. Its length must be equal to number of variables in 'data.x'

KR.corr

When TRUE (default) the extension of the Gower's dissimilarity measure proposed by Kaufman and Rousseeuw (1990) is used. Otherwise, the original Gower's (1971) formula is considered.

na.rm

Replace missing values with maximal distance?

Details

Gower.dist() code based on analogous function from 'StatMatch' package; please see this package for the original code and full documentation.

This function computes the Gower's distance (dissimilarity) among units in a dataset or among observations in two distinct datasets. Columns of mode numeric will be considered as interval scaled variables; columns of mode character or class factor will be considered as categorical nominal variables; columns of class ordered will be considered as categorical ordinal variables and, columns of mode logical will be considered as binary asymmetric variables. Missing values (NA) are allowed. If only data.x is supplied, the dissimilarities between _rows_ of data.x will be computed.

For 'rngs', in correspondence of non-numeric variables, just put 1 or NA. When rngs=NULL (default), the range of a numeric variable is estimated by jointly considering the values for the variable in 'data.x' and those in 'data.y'.

When 'na.rm=TRUE', all missing values (NAs and NaNs) in the result will be replaced with maximal distance. This is discussable but helps, e.g., to bootstrap hierarchical clustering in case if data is rich of NAs.

Value

A distance object with distances among rows of 'data.x' and those of 'data.y'.

Author(s)

Alexey Shipunov

References

Gower J.C. 1971. A general coefficient of similarity and some of its properties. Biometrics. 27: 623–637.

Kaufman L., Rousseeuw P.J. 1990. Finding Groups in Data: An Introduction to Cluster Analysis. Wiley, New York.

See Also

dist, cluster::daisy

Examples

x1 <- as.logical(rbinom(10, 1, 0.5))
x2 <- sample(letters, 10, replace=TRUE)
x3 <- rnorm(10)
x4 <- ordered(cut(x3, -4:4, include.lowest=TRUE))
xx <- data.frame(x1, x2, x3, x4, stringsAsFactors=FALSE)

## matrix of distances among first obs. in xx and the remaining ones
Gower.dist(data.x=xx[1:6, ], data.y=xx[7:10, ])

## matrix of distances among observations in xx
row.names(xx) <- LETTERS[1:nrow(xx)]
dx <- Gower.dist(xx)
plot(hclust(dx))