Whittle likelihood is a statistical method used for estimating parameters in time series models, particularly those involving Gaussian processes and stationary time series. It is named after Peter Whittle, who introduced this likelihood approach. The Whittle likelihood is based on the spectral properties of a time series, specifically its power spectral density (PSD). The key idea is to use the Fourier transform of the data to facilitate parameter estimation.
New to topics? Read the docs here!