Premium
Time‐series forecasting using fractional differencing
Author(s) -
Sutcliffe Andrew
Publication year - 1994
Publication title -
journal of forecasting
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.543
H-Index - 59
eISSN - 1099-131X
pISSN - 0277-6693
DOI - 10.1002/for.3980130404
Subject(s) - series (stratigraphy) , autoregressive integrated moving average , limiting , computer science , integer (computer science) , mathematics , econometrics , long memory , time series , machine learning , geology , paleontology , mechanical engineering , engineering , programming language , volatility (finance)
Abstract The main failure of ARIMA modelling as used in practice are the limiting constraints imposed by differencing to achieve stationarity. The use of fractional differencing opens up a much wider and realistic behaviour for the trend and seasonal components than traditional integer differencing. This paper shows several advantages of using fractional differencing for forecasting monthly data. These advantages are illustrated using a sample of previously modelled time‐series data selected from the literature.