SiZer: SiZer and SiCon plots for 1-dimensional data

Description Usage Arguments Details Value References See Also Examples

View source: R/SiZer.R

Description

SiZer (Significant Zero crossings) and SiCon (Significant Convexity) plots for 1-dimensional data.

Usage

1
2
3
4
5
SiZer(x, bw, gridsize, scaleData=FALSE, signifLevel=0.05, plotSiZer=TRUE,
   logbw=TRUE, xlim, xlab, addLegend=TRUE, posLegend="bottomright") 

SiCon(x, bw, gridsize, scaleData=FALSE, signifLevel=0.05, plotSiCon=TRUE,
   logbw=TRUE, xlim, xlab, addLegend=TRUE, posLegend="bottomright")

Arguments

x

data vector

bw

vector of range of bandwidths

gridsize

number of x- and y-axis grid points

scaleData

flag for scaling the data i.e. transforming to unit variance for each dimension.

signifLevel

significance level

plotSiZer,plotSiCon

flag for displaying SiZer/SiCon map

logbw

flag for displaying log bandwidths on y-axis

xlim

x-axis limits

xlab

x-axis label

addLegend

flag for legend display

posLegend

legend position

Details

The gradient SiZer and curvature SiCon maps of Chaudhuri & Marron (1999) are implemented. The horizontal axis is the data axis, the vertical axis are the bandwidths. The colour scheme for the SiZer map is red: negative gradient, blue: positive gradient, purple: zero gradient and grey: sparse regions. For the SiCon map, orange: negative curvature (concave), blue: positive curvature (convex), green: zero curvature and grey: sparse regions.

Value

SiZer/SiCon plot sent to graphics window.

References

Chaudhuri, P. & Marron, J.S. (1999) SiZer for exploration of structures in curves. Journal of the American Statistical Association, 94, 807-823.

See Also

featureSignif

Examples

1
2
3
4
data(earthquake)
eq3 <- -log10(-earthquake[,3])
SiZer(eq3)
SiCon(eq3)

feature documentation built on Feb. 10, 2021, 9:06 a.m.