Fast ranking function

Source: R/Uhat.R

Function for computing the scaled ranks for each column of the input matrix. In other words, the values are ranked column-wise and divided by nrow(x) + 1. A "1334" ranking scheme is used where the lowest values is awarded rank 1, second lowest value rank 2, and ties are given the maximum available rank.

Uhat(x)

Arguments

x

A numeric matrix of observations to be ranked. Rows correspond to features and columns to experiments.

Value

A matrix with the same dimensions as x of the scaled ranks.

See also

SimulateGMMData, SimulateGMCMData

Author

Anders Ellern Bilgrau <anders.ellern.bilgrau@gmail.com>

Examples

data <- SimulateGMMData()
par(mfrow = c(1,2))
plot(data$z, xlab = expression(z[1]), ylab = expression(z[2]))
plot(Uhat(data$z),
     xlab = expression(hat(u)[1]),
     ylab = expression(hat(u)[2]))