Top-down or bottom-up: Aggregate versus disaggregate extrapolations

By Byron J. Dangerfield, John S. Morris. 1992

How researchers see this paper

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

Golden Rule of Forecasting: Be conservative (2015, methodology): “Dangerfield and Morris (1992) used exponential smoothing models to forecast all 15,753 unique series derived by aggregating pairs of the 178 monthly time-series used in the MCompetition (Makridakis et al., 1982) that included at least 48 observations in the specification set.”
Hierarchical forecasts for Australian domestic tourism (2009, methodology): “, 1968; Edwards and Orcutt, 1969; Kinney, 1971; Dangerfield and Morris, 1992; Zellner and Tobias, 2000) and some finding either method to be uniformly superior (see for example Weatherby, 1984; Fliedner and Mabert, 1992; Shing, 1993). In Section 3 we introduce some notation which neatly generalises ”
Findings from Evidence-Based Forecasting: Methods for Reducing Forecast Error (2006, unknown): “In addition, Dangerfield and Morris (1992), in their study on bottom-up forecasting, found that segmentation was more accurate for 74% of 192 monthly time series from the M-Competition.”