ABSTRACT
This chapter introduces the spectral analysis method as a basic tool for stationary time series analysis. By means of spectral analysis, we can capture the characteristics of time series by decomposing time series into trigonometric functions at each frequency and by representing the features with the magnitude of each periodic component. The spectral analysis method will leads to the definition of the power spectrum and the periodogram of time series, computational methods, variance reduction and smoothing methods. Moreover, an efficient method of computing periodograms is presented using fast Fourier transforms. The chapter considers a method of obtaining an estimator that converges to the true spectrum as the data length n increases.
