The estimation of the frequency, amplitude and phase of a sinusoid from obs
ervations contaminated by correlated noise is considered. It is assumed tha
t the observations are regularly spaced, but may suffer missing values or l
ong time stretches with no data. The typical astronomical source of such da
ta is high-speed photoelectric photometry of pulsating stars. The study of
the observational noise properties of nearly 200 real data sets is reported
: noise can almost always be characterized as a random walk with superposed
white noise. A scheme for obtaining weighted non-linear least-squares esti
mates of the parameters of interest, as well as standard errors of these es
timates, is described. Simulation results are presented for both complete a
nd incomplete data. It is shown that, in finite data sets, results are sens
itive to the initial phase of the sinusoid.