Generalising about univariate forecasting methods: Further empirical evidence

By Robert Fildes, M. Hibon, Spyros Makridakis, Nigel Meade. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Time-series forecasting with deep learning: a survey (2020, background): “]. Prior to the M4 competition of 2018 [61], the prevailing wisdom was that sophisticated methods do not produce more accurate forecasts, and simple models with ensembling had a tendency to do better [59, 62, 63]. Two key reasons have been identiﬁed to explain the underperformance of machine learnin”
25 years of time series forecasting (2006, methodology): “The study of Fildes et al. (1998), which looked at forecasting telecommunications data, used MAPE, MdAPE, PB, AR, GMRAE and MdRAE, taking into account some of the criticism of the methods used for the M-competition.”
Forecasting and operational research: a review (2008, background): “…there is a continuing reliance on ad hoc solutions to overcome outdated perceptions of computer constraints, for example a damaging failure to select optimal smoothing parameters in operations (Fildes et al, 1998) or a limitation to using only 3 years of the total data available for forecasting.”
25 Years of IIF Time Series Forecasting: A Selective Review (2005, methodology): “The study of Fildes et al. (1998), which looked at forecasting telecommunications data, used MAPE, MdAPE, PB, AR, GMRAE and MdRAE, taking into account some of the criticism of the methods used for the M-competition.”
EXPONENTIAL SMOOTHING: THE STATE OF THE ART, PART II (2006, unknown): “In the many re-examinations of the M3 series, Taylor is the only researcher who followed the advice of Fildes (1992) and Fildes et al. (1998) and evaluated forecast performance across time.”
The M3-Competition: results, conclusions and implications (2000, background): “As the competition was run on a (Armstrong & Collopy, 1992, 1993; Makridakis real-time basis the actual state of the economy et al., 1993; Fildes et al., 1998) and have was known to the participating experts, who demonstrated, above any reasonable doubt, the could also find, from published sources, ”
Extrapolation for Time-Series and Cross-Sectional Data (2009, unknown): “Fildes et al. (1998), in a study of 261 telecommunication series, examined forecasts for horizons of one, six, 12, and 18 months, with 1,044 forecasts for each horizon.”
Advances in forecasting with neural networks? Empirical evidence from the NN3 competition on time series prediction (2011, background): “, Fildes, 1992), and extended to provide additional insights (Fildes et al., 1998):”
Findings from Evidence-Based Forecasting: Methods for Reducing Forecast Error (2006, background): “However, Fildes et al. (1998), in a study involving 261 telecommunications series for horizons of up to 12 months, found that damping reduced the accuracy of Holt’s exponential smoothing by 16.8% (based on the two criteria in their table 4); given these unusual findings, an independent replication…”