✦ LIBER ✦

Time-series forecasting using fractional differencing

✍ Scribed by Andrew Sutcliffe

Publisher: John Wiley and Sons
Year: 1994
Tongue: English
Weight: 515 KB
Volume: 13
Category: Article
ISSN: 0277-6693
DOI: 10.1002/for.3980130404

No coin nor oath required. For personal study only.

✦ Synopsis

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 KEY WORDS Fractional differencing ARIMA models Author's biography: Andrew Sutcliffe has an honours degree in mathematics and statistics from Flinders University, South Australia. Currently he is working as a Principal Research Officer at ABARE researching time-series forecasting and decomposition. He has been involved in the application of time-series methods at the Australian Bureau of Statistics and ABARE since 1975.

📜 SIMILAR VOLUMES

Time series forecasting using robust reg

Time series forecasting using robust regression

✍ Hans Levenbach 📂 Article 📅 1982 🏛 John Wiley and Sons 🌐 English ⚖ 710 KB

## Abstract The method of ordinary least squares (OLS) and generalizations of it have been the mainstay of most forecasting methodologies for many years. It is well‐known, however, that outliers or unusual values can have a large influence on least‐squares estimators. Users of automatic forecasting

Forecasting some low-predictability time

Forecasting some low-predictability time series using diffusion indices

✍ Marc Brisson; Bryan Campbell; John W. Galbraith 📂 Article 📅 2003 🏛 John Wiley and Sons 🌐 English ⚖ 254 KB

## Abstract The growth rates of real output and real investment are two macroeconomic time series which are particularly difficult to forecast. This paper considers the application of diffusion index forecasting models to this problem. We begin by characterizing the performance of standard forecast

Forecasting non-normal time series

✍ A. L. Swift; G. J. Janacek 📂 Article 📅 1991 🏛 John Wiley and Sons 🌐 English ⚖ 979 KB

We look at the problem of forecasting time series which are not normally distributed. An overall approach is suggested which works both on simulated data and on real data sets. The idea is intuitively attractive and has the considerable advantage that it can readily be understood by nonspecialists.

Forecasts of the seasonal fractional int

Forecasts of the seasonal fractional integrated series

✍ Olivier Darné; Vivien Guiraud; Michel Terraza 📂 Article 📅 2004 🏛 John Wiley and Sons 🌐 English ⚖ 132 KB

## Abstract We study the probability of rejecting the seasonal unit root tests developed by Hylleberg __et al__. when they are applied to fractionally integrated seasonal time series. We find that these tests have quite low power and that they lead to a risk of over‐differencing. The forecasting pe

Forecasting growth with time series mode

Forecasting growth with time series models

✍ Daniel Peña 📂 Article 📅 1995 🏛 John Wiley and Sons 🌐 English ⚖ 485 KB 👁 1 views

This paper compares the structure of three models for estimating future growth in a time series. It is shown that a regression model gives minimum weight to the last observed growth and maximum weight to the observed growth in the middle of the sample period. A first-order integrated ARIMA model, or

Back propagation in time-series forecast

Back propagation in time-series forecasting

✍ Gerson Lachtermacher; J. David Fuller 📂 Article 📅 1995 🏛 John Wiley and Sons 🌐 English ⚖ 958 KB

## Abstract One of the major constraints on the use of back propagation neural networks as a practical forecasting tool is the number of training patterns needed. We propose a methodology that reduces the data requirements. The general idea is to use the Box‐Jenkins model in an exploratory phase to