wmahalanobis: Compute weighted Mahalanobis distance

In WMDB: Discriminant Analysis Methods by Weight Mahalanobis Distance and bayes

Description

compute weighted Mahalanobis distance between two samples

Usage

wmahalanobis(x, center, cov, weight)

Arguments

x	vector or matrix of data with, say, p columns.
center	mean vector of the distribution or second data vector of length p
cov	covariance matrix (p x p) of the distribution
weight	the weight of the parameters

Details

the weight of parameters is defined by users;if you do not define the weight,the corresponding percent contributions of the parameters based on the principal component analysis scheme will be used instead

Value

weighted Mahalanobis distance will be returned

Author(s)

Bingpei Wu

References

passage: "APPLICATION OF WEIGHTED MAHALANOBIS DISTANCE DISCRIMINANT ANALYSIS METHOD TO CLASSIFICATION OF ROCK MASS QUALITY",whose author is YAO Yinpei

Examples

##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.
x=iris[1:50,1:4]
center=colMeans(x)
cov=var(x)
weight=diag(rep(0.25,4))
wmahalanobis(x,center,cov,weight)

## The function is currently defined as
function (x, center, cov, weight) 
{
    if (is.vector(x)) 
        x = matrix(x, ncol = length(x))
    else x=as.matrix(x)
    x <- sweep(x, 2, center)
    cov <- weight %*% solve(cov)
    retval <- diag(x %*% cov %*% t(x))
    retval
  }

Example output

         1          2          3          4          5          6          7 
0.11227845 0.52027354 0.32108378 0.42655174 0.19042135 0.92816184 0.85604903 
         8          9         10         11         12         13         14 
0.08585981 0.74911912 0.80002148 0.47273815 0.50371984 0.73683328 1.76005248 
        15         16         17         18         19         20         21 
2.55551925 1.91345081 1.43559218 0.15906369 1.29644367 0.40310318 1.33726468 
        22         23         24         25         26         27         28 
0.68058880 2.76110699 1.80759383 2.43699346 0.94262760 0.63142180 0.20728231 
        29         30         31         32         33         34         35 
0.33075371 0.54359893 0.49862653 1.22227788 1.92481960 1.31200598 0.31674524 
        36         37         38         39         40         41         42 
0.82545609 1.43031746 0.77141225 0.81756152 0.14733683 0.42121179 3.08190967 
        43         44         45         46         47         48         49 
1.05025964 3.07751443 2.15028994 0.54866294 0.68893557 0.37221693 0.31318194 
        50 
0.12368898 
function (x, center, cov, weight) 
{
    if (is.vector(x)) 
        x = matrix(x, ncol = length(x))
    else x = as.matrix(x)
    x <- sweep(x, 2, center)
    cov <- weight %*% solve(cov)
    retval <- diag(x %*% cov %*% t(x))
    retval
}

WMDB documentation built on May 2, 2019, 6:12 a.m.