Forecasting time series with multiple seasonal patterns

Gould, Phillip; Koehler, Anne; Ord, J. Keith; Snyder, Ralph D.; Hyndman, Rob J.; Vahid-Araghi, Farshid

doi:10.1016/j.ejor.2007.08.024

Cited by 155 publications

(116 citation statements)

References 14 publications

Supporting

Mentioning

109

Contrasting

Unclassified

Order By: Relevance

“…Other methods [45] are based on this recursive principle: (i) Holt-Winter exponential or double seasonal exponential (see Equations (12) and (13)), (ii) triple exponential (Equations (14)), and the SARIMA (Seasonal ARIMA) presented in [20].…”

Section: Autoregressive Methodsmentioning

confidence: 99%

Interpolation in Time Series: An Introductive Overview of Existing Methods, Their Performance Criteria and Uncertainty Assessment

Lepot

Aubin

Clemens

2017

Water

167

View full text Add to dashboard Cite

Abstract:A thorough review has been performed on interpolation methods to fill gaps in time-series, efficiency criteria, and uncertainty quantifications. On one hand, there are numerous available methods: interpolation, regression, autoregressive, machine learning methods, etc. On the other hand, there are many methods and criteria to estimate efficiencies of these methods, but uncertainties on the interpolated values are rarely calculated. Furthermore, while they are estimated according to standard methods, the prediction uncertainty is not taken into account: a discussion is thus presented on the uncertainty estimation of interpolated/extrapolated data. Finally, some suggestions for further research and a new method are proposed.

show abstract

Section: Autoregressive Methodsmentioning

confidence: 99%

Interpolation in Time Series: An Introductive Overview of Existing Methods, Their Performance Criteria and Uncertainty Assessment

Lepot

Aubin

Clemens

2017

Water

167

View full text Add to dashboard Cite

show abstract

“…The table shows that, for both series, the parameter values are reasonably similar to those of the double seasonal methods. Gould et al (2008) argue that an unappealing feature of the HWT formulation for double seasonality of Section 4.2 is that it assumes that the intraday cycle is the same for each of the seven days of the week. To address this, they allow the intraday cycle for the different days to be represented by different seasonal components.…”

Section: Triple Seasonal Hwt Exponential Smoothingmentioning

confidence: 99%

“…In the same way that we have extended Taylor's (2003) exponential smoothing method for double seasonality, it is straightforward to extend the exponential smoothing model proposed by Gould et al (2008). The triple seasonal formulation is given as: Comparing the parameters for this model with those in Table 1 for the triple seasonal HWT method, we note that the values of λ and φ are similar, while the seasonal component smoothing parameters do not closely obey the conditions for equivalence (γ ii =δ+ω and γ ij =δ) discussed at the end of Section 5.1.…”

Section: Triple Seasonal Ic Exponential Smoothingmentioning

confidence: 99%

“…The focus of Section 4 is exponential smoothing formulations for single, double and triple seasonality based on the Holt-Winters method. In Section 5, the method of Gould et al (2008) is presented, along with its extension for triple seasonality. In Section 6, we describe the neural network that we use as a sophisticated benchmark in our empirical comparison of methods.…”

Section: Introductionmentioning

confidence: 99%

“…In two recent empirical studies, several univariate statistical methods were considered that aim to accommodate both of these seasonal cycles McSharry, 2007, Taylor, 2008). Three of the more successful methods were a double seasonal ARMA model, an adaptation of Holt-Winters exponential smoothing for double seasonality, and an exponential smoothing method recently proposed by Gould et al (2008).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Triple seasonal methods for short-term electricity demand forecasting

Taylor

2010

European Journal of Operational Research

322

229

View full text Add to dashboard Cite

Online short-term load forecasting is needed for the real-time scheduling of electricity generation. Univariate methods have been developed that model the intraweek and intraday seasonal cycles in intraday load data. Three such methods, shown to be competitive in recent empirical studies, are double seasonal ARMA, an adaptation of Holt-Winters exponential smoothing for double seasonality, and another, recently proposed, exponential smoothing method. In multiple years of load data, in addition to intraday and intraweek cycles, an intrayear seasonal cycle is also apparent. We extend the three double seasonal methods in order to accommodate the intrayear seasonal cycle. Using six years of British and French data, we show that for prediction up to a day-ahead the triple seasonal methods outperform the double seasonal methods, and also a univariate neural network approach. Further improvement in accuracy is produced by using a combination of the forecasts from two of the triple seasonal methods.

show abstract

H olt– W inters Exponential Smoothing

Pan

2011

Wiley Encyclopedia of Operations Research and Management Science

View full text Add to dashboard Cite

The Holt–Winters exponential smoothing method was first introduced more than half a century ago for the seasonal time series forecast and it is still one of the most popular forecasting systems widely used in production planning, inventory control, and many other application areas. Numerous empirical studies have confirmed the accuracy and robustness of this simple method, compared to many other sophisticated forecasting techniques. Recently, the underlying optimal time series models of this heuristic method have been found and they provide the theoretical foundation for obtaining the statistical properties of this method. This article aims to provide a succinct summary of the basic Holt–Winters method and help readers understand and appreciate this simple, yet powerful, forecasting technique.

show abstract

Forecasting time series with multiple seasonal patterns

Cited by 155 publications

References 14 publications

Interpolation in Time Series: An Introductive Overview of Existing Methods, Their Performance Criteria and Uncertainty Assessment

Interpolation in Time Series: An Introductive Overview of Existing Methods, Their Performance Criteria and Uncertainty Assessment

Triple seasonal methods for short-term electricity demand forecasting

H olt– W inters Exponential Smoothing

Contact Info

Product

Resources

About