Cameron, Murray A.; Turner, T. Rolf Fitting models to spectra using regression packages. (English) Zbl 0646.62078 J. R. Stat. Soc., Ser. C 36, 47-57 (1987). Summary: Much time series analysis can be performed by using readily available regression packages. In this paper it is shown that several frequency domain estimation algorithms for commonly used time series models may be recast as (possibly iterative) least squares regressions. This approach is exemplified by the fitting of (i) ARMA models, (ii) the model described by P. Bloomfield, Biometrika 60, 217-226 (1973; Zbl 0261.62074), and (iii) a model of A. N. Kolmogorov for the spectrum of turbulence in a fluid. Cited in 2 Documents MSC: 62M15 Inference from stochastic processes and spectral analysis 65C99 Probabilistic methods, stochastic differential equations 62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH) 65K10 Numerical optimization and variational techniques Keywords:autoregressive moving average models; Newton-Raphson; periodogram. regression packages; frequency domain estimation algorithms; time series models; least squares regressions; ARMA models Citations:Zbl 0261.62074 PDFBibTeX XMLCite \textit{M. A. Cameron} and \textit{T. R. Turner}, J. R. Stat. Soc., Ser. C 36, 47--57 (1987; Zbl 0646.62078) Full Text: DOI