A. Montanari et al., FRACTIONALLY DIFFERENCED ARIMA MODELS APPLIED TO HYDROLOGIC TIME-SERIES - IDENTIFICATION, ESTIMATION, AND SIMULATION, Water resources research, 33(5), 1997, pp. 1035-1044
Since Hurst [1951] detected the presence of long-term persistence in h
ydrologic data, new estimation methods and long-memory models have bee
n developed. The lack of flexibility in representing the combined effe
ct of short and long memory has been the major limitation of stochasti
c models used to analyze hydrologic time series. In the present paper
a fractionally differenced autoregressive integrated moving average (F
ARIMA) model is considered. In contrast to using traditional ARIMA mod
els, this approach allows the modeling of both short- and long-term pe
rsistence in a time series. A framework for identification and estimat
ion is presented. The data do not have to be Gaussian. The resulting m
odel, which replicates the sample probability density of the data, can
be used for the generation of long synthetic series. An application t
o the monthly and daily inflows of Lake Maggiore, Italy, is presented.