Improving the forecasting performance of temporal hierarchies

Temporal hierarchies have been widely used during the past few years as they are capable to provide more accurate coherent forecasts at different planning horizons. However, they still display some limitations, being mainly subject to the forecasting methods used for generating the base forecasts and the particularities of the examined series. This paper deals with such limitations by considering three different strategies: (i) combining forecasts of multiple methods, (ii) applying bias adjustments and (iii) selectively implementing temporal hierarchies to avoid seasonal shrinkage. The proposed strategies can be applied either separately or simultaneously, being complements to the method considered for reconciling the base forecasts and completely independent from each other. Their effect is evaluated using the monthly series of the M and M3 competitions. The results are very promising, displaying lots of potential for improving the performance of temporal hierarchies, both in terms of accuracy and bias.

[1]  T. Evgeniou,et al.  To combine or not to combine: selecting among forecasts and their combinations , 2005 .

[2]  Fotios Petropoulos,et al.  Exploring the sources of uncertainty: Why does bagging for time series forecasting work? , 2018, Eur. J. Oper. Res..

[3]  Evangelos Spiliotis,et al.  The M4 Competition: 100,000 time series and 61 forecasting methods , 2020 .

[4]  George Athanasopoulos,et al.  Forecasting: principles and practice , 2013 .

[5]  Amir F. Atiya,et al.  Combination of long term and short term forecasts, with application to tourism demand forecasting , 2011 .

[6]  Yael Grushka-Cockayne,et al.  Trimmed Opinion Pools and the Crowd's Calibration Problem , 2012, Manag. Sci..

[7]  F. Petropoulos,et al.  Improving forecasting by estimating time series structural components across multiple frequencies , 2014 .

[8]  Felix Chan,et al.  Some Theoretical Results on Forecast Combinations , 2015 .

[9]  A. Zellner,et al.  A Note on Aggregation, Disaggregation and Forecasting Performance , 2000 .

[10]  Fotios Petropoulos,et al.  A simple combination of univariate models , 2020 .

[11]  Jeffrey Sohl,et al.  Disaggregation methods to expedite product line forecasting , 1990 .

[12]  Rob J Hyndman,et al.  A state space framework for automatic forecasting using exponential smoothing methods , 2002 .

[13]  Nikolaos Kourentzes,et al.  Neural network ensemble operators for time series forecasting , 2014, Expert Syst. Appl..

[14]  P. Goodwin,et al.  On the asymmetry of the symmetric MAPE , 1999 .

[15]  Philip Hans Franses,et al.  The M3 competition: Statistical tests of the results , 2005 .

[16]  Rob J Hyndman,et al.  Another look at measures of forecast accuracy , 2006 .

[17]  A. Timmermann.  Chapter 4 Forecast Combinations , 2006 .

[18]  Rob J. Hyndman,et al.  FFORMA: Feature-based forecast model averaging , 2020, International Journal of Forecasting.

[19]  F. Petropoulos,et al.  On reconciling macro and micro energy transport forecasts for strategic decision making in the tanker industry , 2017 .

[20]  Nikolaos Kourentzes,et al.  Cross-temporal coherent forecasts for Australian tourism , 2019, Annals of Tourism Research.

[21]  Rob J. Hyndman,et al.  Optimal combination forecasts for hierarchical time series , 2011, Comput. Stat. Data Anal..

[22]  Daniel W. Williams,et al.  Shrinkage estimators of time series seasonal factors and their effect on forecasting accuracy , 2003 .

[23]  Yael Grushka-Cockayne,et al.  Ensembles of Overfit and Overconfident Forecasts , 2015, Management Sciences.

[24]  Robert L. Winkler,et al.  Simple robust averages of forecasts: Some empirical results , 2008 .

[25]  Fotios Petropoulos,et al.  The inventory performance of forecasting methods: Evidence from the M3 competition data , 2019, International Journal of Forecasting.

[26]  Byron J. Dangerfield,et al.  Top-down or bottom-up: Aggregate versus disaggregate extrapolations , 1992 .

[27]  Evangelos Spiliotis,et al.  Forecasting with a hybrid method utilizing data smoothing, a variation of the Theta method and shrinkage of seasonal factors , 2018, International Journal of Production Economics.

[28]  A. Koehler,et al.  Models for optimising the theta method and their relationship to state space models , 2016 .

[29]  Amir F. Atiya,et al.  A Bias and Variance Analysis for Multistep-Ahead Time Series Forecasting , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[30]  Fotios Petropoulos,et al.  forecast: Forecasting functions for time series and linear models , 2018 .

[31]  R. L. Winkler,et al.  Averages of Forecasts: Some Empirical Results , 1983 .

[32]  Gene Fliedner,et al.  An investigation of aggregate variable time series forecast strategies with specific subaggregate time series statistical correlation , 1999, Comput. Oper. Res..

[33]  Fotios Petropoulos,et al.  Forecast combinations for intermittent demand , 2015, J. Oper. Res. Soc..

[34]  Ann Vereecke,et al.  Judgmental forecast adjustments over different time horizons , 2019, Omega.

[35]  Fotios Petropoulos,et al.  Improving forecasting via multiple temporal aggregation , 2014 .

[36]  Pierre Pinson,et al.  Temporal hierarchies with autocorrelation for load forecasting , 2020, Eur. J. Oper. Res..

[37]  Enno Siemsen,et al.  Integrating human judgement into quantitative forecasting methods: A review , 2019, Omega.

[38]  Fotios Petropoulos,et al.  Reconciliation of probabilistic forecasts with an application to wind power , 2018, 1808.02635.

[39]  Hyeongwoo Kim,et al.  Bias Correction and Out-of-Sample Forecast Accuracy , 2009 .

[40]  K. Wallis,et al.  A Simple Explanation of the Forecast Combination Puzzle , 2009 .

[41]  A. Timmermann.  Forecast Combinations , 2005 .

[42]  Andrey L. Vasnev,et al.  The Forecast Combination Puzzle: A Simple Theoretical Explanation , 2014 .

[43]  Nikolaos Kourentzes,et al.  Demand forecasting by temporal aggregation: Using optimal or multiple aggregation levels? , 2017 .

[44]  Helmut Lütkepohl,et al.  Forecasting Contemporaneously Aggregated Vector ARMA Processes , 1984 .

[45]  Rob J Hyndman,et al.  Automatic Time Series Forecasting: The forecast Package for R , 2008 .

[46]  Spyros Makridakis,et al.  The M3-Competition: results, conclusions and implications , 2000 .

[47]  Dipti Srinivasan,et al.  Reconciling solar forecasts: Sequential reconciliation , 2019, Solar Energy.

[48]  Brian Seaman.  Considerations of a retail forecasting practitioner , 2018 .

[49]  Robert L. Winkler,et al.  The accuracy of extrapolation (time series) methods: Results of a forecasting competition , 1982 .

[50]  Fotios Petropoulos,et al.  Forecasting with temporal hierarchies , 2017, Eur. J. Oper. Res..

[51]  Fotios Petropoulos,et al.  Forecasting for big data: Does suboptimality matter? , 2018, Comput. Oper. Res..

[52]  J. M. Bates,et al.  The Combination of Forecasts , 1969 .

[53]  Rob J. Hyndman,et al.  Coherent Probabilistic Forecasts for Hierarchical Time Series , 2017, ICML.

[54]  George Athanasopoulos,et al.  Hierarchical forecasts for Australian domestic tourism , 2009 .

[55]  K. Nikolopoulos,et al.  The theta model: a decomposition approach to forecasting , 2000 .

[56]  Fotios Petropoulos,et al.  Cross-temporal aggregation: Improving the forecast accuracy of hierarchical electricity consumption , 2020, Applied Energy.

[57]  Fotios Petropoulos,et al.  Probabilistic forecast reconciliation with applications to wind power and electric load , 2019, Eur. J. Oper. Res..

[58]  R. Clemen.  Combining forecasts: A review and annotated bibliography , 1989 .

[59]  Fotios Petropoulos,et al.  Forecasting with multivariate temporal aggregation: the case of promotional modelling , 2016 .

[60]  Rob J. Hyndman,et al.  Fast computation of reconciled forecasts for hierarchical and grouped time series , 2016, Comput. Stat. Data Anal..

[61]  Rob J. Hyndman,et al.  Optimal Forecast Reconciliation for Hierarchical and Grouped Time Series Through Trace Minimization , 2018, Journal of the American Statistical Association.