HPWmap: Computation of pointwise and highest pointwise probabilities.

In romanflury/mrbsizeR: Scale Space Multiresolution Analysis of Random Signals

Description

Pointwise (PW) probabilities and highest pointwise (HPW) probabilities of all differences of smooths at neighboring scales are computed.

Usage

HPWmap(smoothVec, mm, nn, prob = 0.95)

Arguments

smoothVec	Differences of smooths at neighboring scales.
mm	Number of rows of the original input image.
nn	Number of columns of the original input image.
prob	Credibility level for the posterior credibility analysis

Details

HPWmap is an internal function of mrbsizeRgrid and is usually not used independently. The output can be analyzed with the plotting function plot.HPWmapGrid.

Value

List with two arrays:

• pw: Pointwise probabilities (VmapPW) including the dimensions of the original input image, mm and nn.

• hpw: Highest pointwise probabilities (VmapHPW) including the dimensions of the original input image, mm and nn.

Examples

# Artificial sample data: 10 observations (5-by-2 object), 10 samples
set.seed(987)
sampleData <- matrix(stats::rnorm(100), nrow = 10)
sampleData[4:6, ] <- sampleData[4:6, ] + 5

# Calculation of the simultaneous credible intervals
HPWmap(smoothVec = sampleData, mm = 5, nn = 2, prob = 0.95)

romanflury/mrbsizeR documentation built on Dec. 15, 2019, 9:30 p.m.