Forecasting: theory and practice - arXiv

1 Forecasting: theory and practiceFotios Petropoulos1, , Daniele Apiletti2, Vassilios Assimakopoulos3, Mohamed Zied Babai4,Devon K. Barrow5, Souhaib Ben Taieb6, Christoph Bergmeir7, Ricardo J. Bessa8, Jakub Bijak9,John E. Boylan10, Jethro Browell11, Claudio Carnevale12, Jennifer L. Castle13, Pasquale Cirillo14,Michael P. Clements15, Clara Cordeiro16,17, Fernando Luiz Cyrino Oliveira18, Shari De Baets19,Alexander Dokumentov20, Joanne Ellison9, Piotr Fiszeder21, Philip Hans Franses22, David T. Frazier23,Michael Gilliland24, M. Sinan G on ul25, Paul Goodwin1, Luigi Grossi26, Yael Grushka-Cockayne27,Mariangela Guidolin26, Massimo Guidolin28, Ulrich Gunter29, Xiaojia Guo30, Renato Guseo26,Nigel Harvey31, David F. Hendry32, Ross Hollyman1, Tim Januschowski33, Jooyoung Jeon34,Victor Richmond R. Jose35, Yanfei Kang36, Anne B. Koehler37, Stephan Kolassa38,10,Nikolaos Kourentzes39,10, Sonia Leva40, Feng Li41, Konstantia Litsiou42, Spyros Makridakis43,Gael M.

2 Martin23, Andrew B. Martinez44,45, Sheik Meeran1, Theodore Modis46,Konstantinos Nikolopoulos47, Dilek Onkal25, Alessia Paccagnini48,49, Anastasios Panagiotelis50,Ioannis Panapakidis51, Jose M. Pav a52, Manuela Pedio53,54, Diego J. Pedregal55, Pierre Pinson56,Patr cia Ramos57, David E. Rapach58, J. James Reade59, Bahman Rostami-Tabar60, Micha Rubaszek61,Georgios Sermpinis62, Han Lin Shang63, Evangelos Spiliotis3, Aris A. Syntetos60,Priyanga Dilini Talagala64, Thiyanga S. Talagala65, Len Tashman66, Dimitrios Thomakos67,Thordis Thorarinsdottir68, Ezio Todini69,70, Juan Ram on Trapero Arenas55, Xiaoqian Wang36,Robert L. Winkler71, Alisa Yusupova10, Florian Ziel721 School of Management, University of Bath, UK2 Politecnico di Torino, Turin, Italy3 Forecasting and Strategy Unit, School of Electrical and ComputerEngineering, National Technical University of Athens, Greece4 Kedge Business School, France5 Department of Management, Birmingham Business School,University of Birmingham, UK6 Big Data and Machine Learning Lab, Universit e de Mons (UMONS)

3 , Belgium7 Faculty of Information Technology, MonashUniversity, Melbourne, Australia8 INESC TEC Institute for Systems and Computer Engineering, Technology and Science, Porto, Portugal9 Department ofSocial Statistics and Demography, University of Southampton, UK10 Centre for Marketing Analytics and Forecasting, Lancaster University ManagementSchool, Lancaster University, UK11 School of Mathematics and Statistics, University of Glasgow, UK12 Department of Mechanical and IndustrialEngineering, University of Brescia, Italy13 Magdalen College, University of Oxford, UK14 ZHAW School of Management and Law, Zurich University ofApplied Sciences, Switzerland15 ICMA Centre, Henley Business School, University of Reading, UK16 Faculdade de Ci encias e Tecnologia, Universidade doAlgarve, Portugal17 CEAUL, Faculdade de Ci encias, Universidade de Lisboa, Portugal18 Pontifical Catholic University of Rio de Janeiro (PUC-Rio)

4 , Brazil19 Department of Business Informatics and Operations Management, Faculty of Economics and Business Administration, Universiteit Gent, Belgium20 Let sForecast, Australia21 Faculty of Economic Sciences and Management, Nicolaus Copernicus University in Torun, Poland22 Econometric Institute, ErasmusSchool of Economics, Rotterdam, The Netherlands23 Department of Econometrics and Business Statistics, Monash University, Melbourne, Australia24 SAS,USA25 Newcastle Business School, Northumbria University, Newcastle upon Tyne, UK26 Department of Statistical Sciences, University of Padua, Italy27 Darden School of Business, University of Virginia, USA28 Finance Department, Bocconi University and Baffi-CAREFIN Centre, Milan, Italy29 Departmentof Tourism and Service Management, MODUL University Vienna, Austria30 Robert H. Smith School of Business, University of Maryland, USA31 Department of Experimental Psychology, University College London, UK32 Nuffield College and Institute for New Economic Thinking at the Oxford MartinSchool, University of Oxford, UK33 Amazon Research, Germany34 Korea Advanced Institute of Science and Technology, South Korea35 McDonough Schoolof Business, Georgetown University, USA36 School of Economics and Management, Beihang University, Beijing, China37 Miami University, Ohio, USA38 SAP, Switzerland39Sk ovde Artificial Intelligence Lab, School of Informatics, University of Sk ovde, Sweden40 Department of Energy, Politecnico di Milano,Italy41 School of Statistics and Mathematics, Central University of Finance and Economics, Beijing, China42 Manchester Metropolitan University BusinessSchool.

5 UK43M Open Forecasting Center & Institute for the Future, University of Nicosia, Nicosia, Cyprus44 Office of Macroeconomic Analysis, USDepartment of the Treasury, Washington DC, USA45 GWU Research Program on Forecasting, Washington DC, USA46 Growth Dynamics, Lugano,Switzerland47 Durham University Business School, Durham University, UK48 Michael Smurfit Business School, University College Dublin, Ireland49 Centre for Applied Macroeconomic Analysis, Australia50 Discipline of Business Analytics, The University of Sydney Business School, Australia51 Department of Electrical and Computer Engineering, University of Thessaly, Volos, Greece52 GIPEyOP, UMMICS, Department of Applied Economics,Universitat de Valencia, Spain53 School of Accounting and Finance, University of Bristol, UK54 Baffi-CAREFIN Centre, Bocconi University, Italy55 ETSII ndustrial, Universidad de Castilla-La Mancha, Ciudad Real, Spain56 Department of Technology, Management and Economics, Technical University ofDenmark, Denmark57 Porto Accounting and Business School, Polytechnic of Porto, Portugal58 Department of Economics, Saint Louis University, USA59 Department of Economics, School of Politics, Economics and International Relations, University of Reading, UK60 CardiffBusiness School, CardiffUniversity, UK61 SGH Warsaw School of Economics, Collegium of Economic Analysis, Poland62 Adam Smith Business School, University of Glasgow, UK63 Department of Actuarial Studies and Business Analytics, Macquarie University, Australia64 Department of Computational Mathematics, University ofMoratuwa, Sri Lanka65 Department of Statistics, Faculty of Applied Sciences, University of Sri Jayewardenepura.

6 Sri Lanka66 Foresight, InternationalInstitute of Forecasters, USA67 School of Economics and Political Science, National and Kapodistrian University of Athens, Greece68 Norwegian ComputingCenter, Oslo, Norway69 University of Bologna, Italy70 Italian Hydrological Society, Bologna, Italy71 Fuqua School of Business, Duke University, Durham,USA72 House of Energy Markets and Finance, University of Duisburg-Essen, GermanyPreprint submitted to International Journal of ForecastingJanuary 7, 2022 [ ] 5 Jan 2022In theory , there is no difference between theoryand practice . But, in practice , there Brewster (1882)AbstractForecasting has always been at the forefront of decision making and planning. The uncertainty thatsurrounds the future is both exciting and challenging, with individuals and organisations seeking to min-imise risks and maximise utilities.

7 The large number of forecasting applications calls for a diverse set offorecasting methods to tackle real-life challenges. This article provides a non-systematic review of the the-ory and the practice of forecasting. We provide an overview of a wide range of theoretical, state-of-the-artmodels, methods, principles, and approaches to prepare, produce, organise, and evaluate forecasts. Wethen demonstrate how such theoretical concepts are applied in a variety of real-life do not claim that this review is an exhaustive list of methods and applications. However, we wishthat our encyclopedic presentation will offer a point of reference for the rich work that has been undertakenover the last decades, with some key insights for the future of forecasting theory and practice . Given itsencyclopedic nature, the intended mode of reading is non-linear.

8 We offer cross-references to allow thereaders to navigate through the various topics. We complement the theoretical concepts and applicationscovered by large lists of free or open-source software implementations and publicly-available :review; encyclopedia; methods; applications; principles; time series; prediction. Corresponding author: Introduction (Robert L. Winkler)72 Introduction to forecasting theory (Anne B. Koehler) .. Pre-processing data .. transformations (Anastasios Panagiotelis) .. series decomposition (Alexander Dokumentov) .. detection and time series forecasting (Priyanga Dilini Talagala) .. handling of outliers in time series forecasting (Luigi Grossi) .. variables and feature engineering (Jethro Browell) .. Statistical and econometric models .. smoothing models (Juan Ram on Trapero Arenas).

9 Regression models (Vassilios Assimakopoulos) .. method and models (Dimitrios Thomakos) .. integrated moving average (ARIMA) models (Philip Hans Franses & Sheik Meeran) for multiple seasonal cycles (Bahman Rostami-Tabar) .. models (Diego J. Pedregal) .. for population processes (Jakub Bijak) .. count time series (Gael M. Martin) .. with many variables (J. James Reade) .. Functional time series models (Han Lin Shang) .. ARCH/GARCH models (Jooyoung Jeon) .. Markov switching models (Massimo Guidolin) .. Threshold models (Manuela Pedio) .. Low and high prices in volatility models (Piotr Fiszeder) .. Forecasting with DSGE models (Alessia Paccagnini) .. Robust equilibrium-correction forecasting devices (Andrew B. Martinez) .. Forecasting with data subject to revision (Michael P.)

10 Clements) .. Innovation diffusion models (Mariangela Guidolin) .. The natural law of growth in competition (Theodore Modis) .. Synchronic and diachronic competition (Renato Guseo) .. Estimation and representation of uncertainty (Ricardo Bessa) .. Forecasting under fat tails (Pasquale Cirillo) .. Bayesian forecasting .. of Bayesian forecasting (David T. Frazier & Gael M. Martin) .. of Bayesian forecasting (David T. Frazier & Gael M. Martin) .. forecasting with copulas (Feng Li) .. Variable and model selection .. indicators and Granger causality (Ulrich Gunter) .. complexity (Micha Rubaszek) .. selection (Ross Hollyman) .. selection (David F. Hendry) .. for time-series data (Christoph Bergmeir) .. Combining forecasts .. combination: a brief review of statistical approaches (Devon K.