FitHReg: Fits Three Parameter Harmonic Regression

In pRSR: Test of Periodicity using Response Surface Regression

Description

Estimates A, B and f in the harmonic regression, y(t)=mu+A*cos(2*pi*f*t)+B*sin(2*pi*f*t)+e(t) using LS.

Usage

FitHReg(y, t = 1:length(y), nf=150)

Arguments

y	series
t	time points
nf	nf, number of frequencies to enumerate

Details

Program is interfaced to C for efficient computation.

Value

Object of class "HReg" produced. This is a list with components: 'coefficients', 'residuals', 'Rsq', 'fstatistic', 'sigma', 'freq', 'LRStat' corresponding to the 3 regression coefficients, residuals, R-squared, F-statistic, residual sd, optimal frequency and LR-test statistic for null hypothesis white noise.

References

Islam, M.S. (2008). Peridocity, Change Detection and Prediction in Microarrays. Ph.D. Thesis, The University of Western Ontario.

See Also

GetFitHReg

Examples

z<-SimulateHReg(10, f=2.5/10, 1, 2)
FitHReg(z)

pRSR documentation built on May 2, 2019, 12:09 a.m.