CVST: noisyDonoho – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

noisyDonoho

Generate Donoho's Toy Data Sets

Description

This function allows to generate noisy variants of the toy signals introduced by Donoho (see reference section). The scaling is chosen to reflect the setting as discussed in the original paper.

Usage

noisyDonoho(n, fun = doppler, sigma = 1)
blocks(x, scale = 3.656993)
bumps(x, scale = 10.52884)
doppler(x, scale = 24.22172)
heavisine(x, scale = 2.356934)

Arguments

`n`	Number of data points that should be generated.
`fun`	Function to use to generate the data.
`sigma`	Standard deviation of the noise component.
`x`	Number of data points that should be generated.
`scale`	Scaling parameter.

Value

Returns a data set of type CVST.data

Author(s)

Tammo Krueger <tammokrueger@googlemail.com>

References

David L. Donoho and Jain M. Johnstone. Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81 (3) 425–455, 1994.

Examples

bumpsSet = noisyDonoho(1000, fun=bumps)
plot(bumpsSet)
dopplerSet = noisyDonoho(1000, fun=doppler)
plot(dopplerSet)

CVST

Fast Cross-Validation via Sequential Testing

v0.2-2

GPL (>= 2.0)

Authors

Tammo Krueger, Mikio Braun

Initial release

2018-05-26