It's time to move from 'what' to 'why'

By Rob J Hyndman. 2001

32 citations recorded by the source corpus.

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

Optimal combination forecasts for hierarchical time series (2011, methodology): “For each series, we select an innovations state space model based on Hyndman et al. (2002) using all of the available data.”
Automatic Time Series Forecasting: The forecast Package for R (2008, methodology): “(See Hyndman and Kostenko (2007) for the minimal sample size required.) The smoothing parameters are optimized by minimizing the average squared prediction errors, which is equivalent to minimizing (7) in the case of additive errors. There is a predict method for the resulting object which can produ”
EXPONENTIAL SMOOTHING: THE STATE OF THE ART, PART II (2006, unknown): “Unfortunately, maximum likelihood may require significant computation times, as discussed in Hyndman et al. (2002). For example, in monthly seasonal models with a damped trend, there are 13 initial values and 4 parameters, so the optimization is done in 17-dimensional space.”
25 Years of IIF Time Series Forecasting: A Selective Review (2005, methodology): “In a simulation study, Hyndman (2001) showed that exponential smoothing performs better than ARIMA models because it is not so subject to model selection problems, particularly when data are non-normal.”
A brief history of forecasting competitions (2020, background): “• evaluating prediction intervals (Goodrich 2001); • including higher frequency data such as weekly and daily data (Goodrich 2001); • evaluating multivariate forecasting models (Granger 2001); • evaluating the reasons behind the differences between methods (Hyndman 2001), and • identifying the circu”
25 years of time series forecasting (2006, methodology): “However, Grunwald, Hamza & Hyndman (1997) showed that most of the commonly used models have the substantial flaw of all sample paths converging to a constant when the sample space is less than the whole real line, making them unsuitable for anything other than point forecasting.”