A function of an angular frequency defining a weight function used in the non-parametric estimation of the spectral density of a stationary stochastic process by smoothing the periodogram constructed from the observed data of the process. As an estimator of the value of the spectral density at a point one usually takes the integral with respect to of the product of the periodogram at and an expression of the form . Here is a real number and is a fixed function of the frequency which takes its greatest value at and is such that its integral over is equal to one. This function is usually called a spectral window generator, while the term "spectral window" is used for the function . The width of the spectral window is , and depends on the size of the sample (that is, on the length of the observed realization of the process ) and tends to zero as (but more slowly than ). The Fourier transform of the spectral window (and in the case of discrete time , when , the set of its Fourier coefficients) is called the lag window of an estimator of the spectral density. It defines a weight function of a discrete or continuous argument (depending on whether is discrete or continuous), by which one must multiply the sample auto-correlations evaluated from the given sample to make the Fourier transform of the resulting product coincide with the desired estimator of the spectral density (cf. Spectral density, estimator of the).