Representation of observational data with infinite correlation time: The failure of autoregressive modeling

In short observational records, with reducing the data length it becomes increasingly difficult to distinguish between long-term and short-term memory based on the analysis of single data series. Here we investigate the applicability of autoregressive modelling to long-term correlated data. The applicability of AR model is shown in terms of the autocorrelation function and power spectral density.