mrl: Mean Residual Life Plot

In RecordLinkage: Record Linkage Functions for Linking and Deduplicating Data Sets

Mean Residual Life Plot

Description

Create mean residual life statistics for classification based on extreme value theory.

Usage

mrl(data, umin = min(data), umax = max(data) - 0.1, 
  nint = round(max(data) - min(data)) * 20)

plotMRL(rpairs, l = .computeMRL(sort(as.ram((rpairs$Wdata)))))

Arguments

data	A numerical vector, representing weights in the context of Record Linkage.
umin, umax	The minimum and maximum thresholds at which the mean residual life function is calculated.
nint	The number of points at which the mean residual life function is calculated.
rpairs	A RecLinkData object with assigned weights.
l	A list with components x and y representing MRL statistics as returned by mrl.

Details

While mrl only returns MRL statistics as a list, plotMRL produces a plot thereof. These functions are used internally for threshold estimation based on Extreme Value Theory, see getParetoThreshold.

mrl is a simplified version of mrl.plot in package ismev.

Value

mrl returns a list with

x	x-coordinates of MRL statistics
y	y-coordinates of MRL statistics

See Also

getParetoThreshold

RecordLinkage documentation built on Nov. 10, 2022, 5:42 p.m.