Statistical estimation of nonstationary Gaussian processes with long-range dependence and intermittency. (Q1766082)

A nonstationary Gaussian process with stationary increments is considered; the spectral density of the increments process is supposed to have a general and flexible form which concludes the spectral density of the fractional Riesz-Bessel motion as a special case. A continuous version of the periodogram is used to construct (Gauss-Whittle) estimation procedure for the parameters of the spectral density function. Strong consistency and asymptotic normality of corresponding estimators are established. The procedure is applied to real data from the air pollution monitoring.