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
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.
Details
Program is interfaced to C for efficient computation.
References
Islam, M.S. (2008).
Peridocity, Change Detection and Prediction in Microarrays.
Ph.D. Thesis, The University of Western Ontario.