denoise.chisq

denoise.chisq

denoise.chisq(x, denoise.gauss = f.u, M = logb(length(x), 2), cycles = 10)

Denoises sequences of (possibly correlated) chi-squares using the Haar-Fisz transform. If denoise.gauss is set to f.u, then WaveThresh3 is required.

Takes:

x The sequence of chi-squares, its length must be a power of 2.

denoise.gauss Unquoted name of an S-Plus routine for denoising vectors contaminated with stationary Gaussian noise. Must take and return a vector of length 2^J where J is an integer. The following routines supplied in this package can be used here: f.u, f.sp. The user can define and plug in his or her own routines here.

M The resolution of the Haar-Fisz transform used in the denoising algorithm, see Section 5.3.1 of the thesis.

cycles The number of external cycle spins to be performed.

Returns:

est Denoised version of x (a vector of the same length as x).

x	The sequence of chi-squares, its length must be a power of 2.
denoise.gauss	Unquoted name of an S-Plus routine for denoising vectors contaminated with stationary Gaussian noise. Must take and return a vector of length 2^J where J is an integer. The following routines supplied in this package can be used here: f.u, f.sp. The user can define and plug in his or her own routines here.
M	The resolution of the Haar-Fisz transform used in the denoising algorithm, see Section 5.3.1 of the thesis.
cycles	The number of external cycle spins to be performed.