Souhaib Ben Taieb, James W. Taylor, Rob J. Hyndman

Many applications require forecasts for a hierarchy comprising a set of time series along with aggregates of subsets of these series. Although forecasts can be produced independently for each series in the hierarchy, typically this does not lead to coherent forecasts — the property that forecasts add up appropriately across the hierarchy. State-of-the-art hierarchical forecasting methods usually reconcile these independently generated forecasts to satisfy the aggregation constraints. A fundamental limitation of prior research is that it has looked only at the problem of forecasting the mean of each time series. We consider the situation where probabilistic forecasts are needed for each series in the hierarchy. We define forecast coherency in this setting, and propose an algorithm to compute predictive distributions for each series in the hierarchy. Our algorithm has the advantage of synthesizing information from different levels in the hierarchy through a sparse forecast combination and a probabilistic hierarchical aggregation. We evaluate the accuracy of our forecasting algorithm on both simulated data and large-scale electricity smart meter data. The results show consistent performance gains compared to state-of-the art methods

Download working paper

  Tag: forecasting

1 2 3
March 7th, 2017

Coherent Probabilistic Forecasts for Hierarchical Time Series

Souhaib Ben Taieb, James W. Taylor, Rob J. Hyndman Abstract: Many applications require forecasts for a hierarchy comprising a set […]

February 28th, 2017

Forecasting with temporal hierarchies

George Athanasopoulosa, Rob J. Hyndmana, Nikolaos Kourentzesb, Fotios Petropoulosc a Department of Econometrics and Business Statistics, Monash University, Australia b […]

February 14th, 2017

Macroeconomic forecasting for Australia using a large number of predictors

Bin Jiang, George Athanasopoulos, Rob J Hyndman, Anastasios Panagiotelis, Farshid Vahid Abstract A popular approach to forecasting macroeconomic variables is […]

January 13th, 2017

Visualising forecasting algorithm performance using time series instance spaces

Yanfei Kang1, Rob J Hyndman2, Kate Smith-Miles3 School of Statistics, Renmin University of China. Department of Econometrics and Business Statistics, […]

January 1st, 2017

Grouped functional time series forecasting: an application to age-specific mortality rates

Han Lin Shang and Rob J Hyndman Journal of Computational and Graphical Statistics (2017) to appear. Abstract Age-specific mortality rates […]

October 13th, 2016

Reconciling forecasts: the hts and thief packages

Talk given at eRum2016, Poznań, Poland.  

September 15th, 2016

Forecasting large collections of related time series

Talk given at the German Statistical Week, Augsburg, 15 September 2016

August 22nd, 2016

thief package for R

The thief package provides tools for Temporal Hierarchical Forecasting. The methods are described in Forecasting with temporal hierarchies, co-authored with […]

May 25th, 2016

ISCRR time series workshop

Materials for one-day workshop on time series and forecasting presented to ISCRR, Melbourne. Textbook Hyndman and Athanasopoulos (2014): Software […]

May 6th, 2016

Automatic foRecasting using R

Talk given at the Melbourne Data Science Initiative, 6 May 2016.  

February 4th, 2016

Forecasting uncertainty in electricity smart meter data by boosting additive quantile regression

Souhaib BenTaieb, Raphael Huser, Rob J. Hyndman and Marc G. Genton IEEE Transactions on Smart Grid (2016), 7(5), 2448-2455. Abstract: […]

January 25th, 2016

Probabilistic Energy Forecasting: Global Energy Forecasting Competition 2014 and Beyond

Tao Hong, Pierre Pinson, Shu Fan, Hamidreza Zareipour, Alberto Troccoli and Rob J Hyndman International Journal of Forecasting (2016), 32(3), 896–913. […]

January 24th, 2016

Long-term forecasts of age-specific participation rates with functional data models

Thomas Url1, Rob J Hyndman2, Alexander Dokumentov2 Vienna University of Economics and Business, Vienna, Austria Monash Business School, Monash University, […]

January 1st, 2016

Bagging exponential smoothing methods using STL decomposition and Box-Cox transformation

Christoph Bergmeir1, Rob J Hyndman2, José M Benítez1 Department of Computer Science and Artificial Intelligence, University of Granada, Spain. Department […]

January 1st, 2016

Fast computation of reconciled forecasts for hierarchical and grouped time series

By Rob J Hyndman, Ala Lee & Earo Wang

December 31st, 2015

Measuring forecast accuracy

This is a chapter for a new book on forecasting in business.

November 26th, 2015

Forecasting hierarchical and grouped time series through trace minimization

Shanika L Wickramasuriya, George Athanasopoulos, Rob J Hyndman Department of Econometrics and Business Statistics, Monash University   Abstract Large collections […]

November 2nd, 2015

Forecasting big time series data using R

Keynote address given at the Chinese R conference held in Nanchang, Jianxi province. 24-25 October 2015.

August 17th, 2015

Machine learning bootcamp

A talk on time series forecasting for the Monash University Machine Learning Bootcamp. Demo R code

June 23rd, 2015

MEFM: An R package for long-term probabilistic forecasting of electricity demand

International Symposium on Forecasting Riverside, California   I will describe and demonstrate a new open-source R package that implements the […]

June 19th, 2015

Probabilistic forecasting of peak electricity demand

Southern California Edison Rosemead, California   Electricity demand forecasting plays an important role in short-term load allocation and long-term planning […]

June 8th, 2015

STR: A Seasonal-Trend Decomposition Procedure Based on Regression

By Alex Dokumentov and Rob J Hyndman

June 4th, 2015

Probabilistic time series forecasting with boosted additive models: an application to smart meter data

By Souhaib Ben Taieb, Raphael Huser, Rob J Hyndman and Marc G Genton

May 26th, 2015

Visualization of big time series data

Talk given to a joint meeting of the Statistical Society of Australia (Victorian branch) and the Melbourne Data Science Meetup Group.

May 22nd, 2015

Probabilistic forecasting of long-term peak electricity demand

The latest version of my talk on electricity demand forecasting.
Given to the “Monash Energy Materials and Systems Institute”

April 20th, 2015

A note on the validity of cross-validation for evaluating time series prediction

By Christoph Bergmeir, Rob J Hyndman, and Bonsoo Koo

April 4th, 2015

Discussion of “High-dimensional autocovariance matrices and optimal linear prediction”

My discussion of the article on “High-dimensional autocovariance matrices and optimal linear prediction”
by Timothy L. McMurry and Dimitris N. Politis.

Electronic J Statistics (2015) 9, 792-796.

February 23rd, 2015

Visualization and forecasting of big time series data

Talk given at the ACEMS Big data workshop, QUT.

January 12th, 2015

Visualizing and forecasting big time series data

Talk given at the Institute of Statistical Science, Academia Sinica, Taipei, Taiwan.

December 17th, 2014

MEFM package for R

The MEFM package for R includes a set of tools for implementing the Monash Electricity Forecasting Model.

October 21st, 2014

Optimally reconciling forecasts in a hierarchy

By Rob J Hyndman and George Athanasopoulos

Foresight (Fall, 2014). pp.42-48.

September 23rd, 2014

Forecasting: principles and practice (UWA course)

Workshop held at UWA on 23-25 September 2014.

July 1st, 2014

Fast computation of reconciled forecasts in hierarchical and grouped time series

Talk given at the International Symposium on Forecasting, Rotterdam.

June 17th, 2014

Challenges in forecasting peak electricity demand

A two-part seminar given at the Energy Forum, Valais/Wallis, Switzerland.

June 5th, 2014

Low-dimensional decomposition, smoothing and forecasting of sparse functional data

By Alexander Dokumentov and Rob J Hyndman

May 30th, 2014

State space models

A one-day workshop for the Australian Bureau of Statistics

May 22nd, 2014

Monash Electricity Forecasting Model

By Rob J Hyndman and Shu Fan

May 8th, 2014

forecast package for R

The forecast package for R provides methods and tools for displaying and analysing univariate time series forecasts including exponential smoothing via state space models and automatic ARIMA modelling.

April 9th, 2014

hts package for R

The hts package provides methods for analysing and forecasting hierarchical time series.

April 1st, 2014

A gradient boosting approach to the Kaggle load forecasting competition

By Souhaib Ben Taieb and Rob J Hyndman

International Journal of Forecasting (2014), 30(2), 382–394.

January 10th, 2014

Boosting multi-step autoregressive forecasts

Souhaib Ben Taieb and Rob J Hyndman International Conference on Machine Learning (ICML) 2014. Abstract Multi-step forecasts can be produced […]

January 1st, 2014

Forecasting: principles and practice


October 10th, 2013

Forecasting hierarchical time series

Talk given at University of Sydney today.

June 25th, 2013

Forecasting without forecasters

A keynote talk given at the International Symposium on Forecasting, Seoul, South Korea. 25 June 2013.

June 19th, 2013

Mcomp package for R

The Mcomp package for R provides the 1001 time series from the M-competition (Makridakis et al. 1982) and the 3003 […]

March 14th, 2013

fpp package for R

The fpp package for R provides all data sets required for the examples and exercises in the book Forecasting: principles and […]

January 1st, 2013

A change of editors

International Journal of Forecasting (2013) 29(1), page A1.

October 30th, 2012

expsmooth package for R

The expsmooth package for R provides data sets from the book “Forecasting with exponential smoothing: the state space approach” by […]

October 30th, 2012

fma package for R

The fma package for R provides all data sets from “Forecasting: methods and applications” by Makridakis, Wheelwright & Hyndman (Wiley, […]

September 2nd, 2012

Recursive and direct multi-step forecasting: the best of both worlds

Souhaib Ben Taieb1 and Rob J Hyndman2 Université Libre de Bruxelles Monash University Abstract: We propose a new forecasting strategy, […]