Filters








776 Hits in 7.0 sec

Approximate Bayesian Inference in Linear State Space Models for Intermittent Demand Forecasting at Scale [article]

Matthias Seeger, Syama Rangapuram, Yuyang Wang, David Salinas, Jan Gasthaus, Tim Januschowski, Valentin Flunkert
2017 arXiv   pre-print
We present a scalable and robust Bayesian inference method for linear state space models.  ...  The method is applied to demand forecasting in the context of a large e-commerce platform, paying special attention to intermittent and bursty target statistics.  ...  Latent State Forecasting In this section, we develop latent state forecasting for intermittent demand, combining GLMs, general likelihoods, and exponential smoothing time series models.  ... 
arXiv:1709.07638v1 fatcat:vu4m25wxa5dr5ggsumjllygbqm

Bayesian Intermittent Demand Forecasting for Large Inventories

Matthias W. Seeger, David Salinas, Valentin Flunkert
2016 Neural Information Processing Systems  
We present a scalable and robust Bayesian method for demand forecasting in the context of a large e-commerce platform, paying special attention to intermittent and bursty target statistics.  ...  Inference is approximated by the Newton-Raphson algorithm, reduced to linear-time Kalman smoothing, which allows us to operate on several orders of magnitude larger problems than previous related work.  ...  Acknowledgements We would like to thank Maren Mahsereci for determining the running time figures, and the Wupper team for all the hard work without which this paper would not have happened.  ... 
dblp:conf/nips/SeegerSF16 fatcat:dguscyowqng25bjjegfm6xtuey

Effective Bayesian Modeling of Groups of Related Count Time Series [article]

Nicolas Chapados
2014 arXiv   pre-print
Time series of counts arise in a variety of forecasting applications, for which traditional models are generally inappropriate.  ...  We derive an efficient approximate inference technique, and illustrate its performance on a number of datasets from supply chain planning.  ...  Acknowledgments The author wishes to thank his colleagues at ApSTAT Technologies and JDA Software for constructive discussions, as well as the anonymous reviewers for their insight and useful comments.  ... 
arXiv:1405.3738v1 fatcat:xce2nscxwzdhzfv7fevmblkzd4

Bayesian forecasting of many count-valued time series [article]

Lindsay Berry, Mike West
2018 arXiv   pre-print
Novel univariate models synthesise dynamic generalized linear models for binary and conditionally Poisson time series, with dynamic random effects for over-dispersion.  ...  New multivariate models then enable information sharing in contexts when data at a more highly aggregated level provide more incisive inferences on shared patterns such as trends and seasonality.  ...  There are many benefits to Bayesian state-space modeling in this context.  ... 
arXiv:1805.05232v1 fatcat:x4mg3nxsobdifiybseklk62w7e

Probabilistic forecasting of heterogeneous consumer transaction-sales time series [article]

Lindsay R. Berry, Paul Helman, Mike West
2018 arXiv   pre-print
Keywords: Bayesian forecasting; decouple/recouple; dynamic binary cascade; forecast calibration; intermittent demand; multi-scale forecasting; predicting rare events; sales per transaction; supermarket  ...  We present new Bayesian methodology for consumer sales forecasting.  ...  We acknowledge discussions and data development with Xiaojie Zhou and others in the research team at 84.51 • .  ... 
arXiv:1808.04698v2 fatcat:tpbeemsi7ze7bdfphgvd3lbkfa

Deep Factors for Forecasting [article]

Yuyang Wang, Alex Smola, Danielle C. Maddix, Jan Gasthaus, Dean Foster, Tim Januschowski
2019 arXiv   pre-print
Classical time series models fail to capture complex patterns in the data, and multivariate techniques struggle to scale to large problem sizes.  ...  Producing probabilistic forecasts for large collections of similar and/or dependent time series is a practically relevant and challenging task.  ...  Bayesian Kingma, D. P. and Welling, M. Auto-encoding variational intermittent demand forecasting for large inventories. In bayes.  ... 
arXiv:1905.12417v1 fatcat:6jzizdhebfb43goezaplgsonka

Forecasting: theory and practice [article]

Fotios Petropoulos, Daniele Apiletti, Vassilios Assimakopoulos, Mohamed Zied Babai, Devon K. Barrow, Souhaib Ben Taieb, Christoph Bergmeir, Ricardo J. Bessa, Jakub Bijak, John E. Boylan, Jethro Browell, Claudio Carnevale (+68 others)
2022 arXiv   pre-print
We provide an overview of a wide range of theoretical, state-of-the-art models, methods, principles, and approaches to prepare, produce, organise, and evaluate forecasts.  ...  Forecasting has always been at the forefront of decision making and planning.  ...  State-space models 13 State Space (SS) systems are a very powerful and useful framework for time series and econometric modelling and forecasting.  ... 
arXiv:2012.03854v4 fatcat:p32c67sy65cfdejq7ndfs3g7dm

State space models for non‐stationary intermittently coupled systems: an application to the North Atlantic oscillation

Philip G. Sansom, Daniel B. Williamson, David B. Stephenson
2019 Journal of the Royal Statistical Society, Series C: Applied Statistics  
We develop Bayesian state space methods for modelling changes to the mean level or temporal correlation structure of an observed time series due to intermittent coupling with an unobserved process.  ...  Skilful forecasts for the winter (December-January-February) mean are possible from the beginning of December.  ...  We also thank two reviewers and the Associate Editor for their helpful comments.  ... 
doi:10.1111/rssc.12354 fatcat:ocjqr6mnwjbnfewu62rj3fwl6q

State space models for non-stationary intermittently coupled systems: an application to the North Atlantic Oscillation [article]

Philip G. Sansom, and Daniel B. Williamson, David B. Stephenson
2019 arXiv   pre-print
We develop Bayesian state space methods for modelling changes to the mean level or temporal correlation structure of an observed time series due to intermittent coupling with an unobserved process.  ...  Skilful forecasts for winter (Dec-Jan-Feb) mean are possible from the beginning of December.  ...  Discussion In this study we have developed Bayesian state space methods for diagnosing predictability in intermittently coupled systems.  ... 
arXiv:1711.04135v3 fatcat:7dzbxptu2raixgfom4fdzxqgjy

Integration of Judgmental and Statistical Approaches for Demand Forecasting: Models and Methods

Andrey Davydenko
2020 figshare.com  
The aim of the research is to develop efficient models and methods which would better correspond to realistic problem definitions in the context of demand forecasting.  ...  However, due to the specific nature of demand data existing solutions in this area often cannot be efficiently applied in demand forecasting.  ...  Product demand forecasts and corresponding 80% prediction intervalsobtained at different points using exponential smoothing state space models (a,b,c) and the proposed approach (d,e,f).  ... 
doi:10.6084/m9.figshare.13513212.v1 fatcat:kg7z2db3ufgcte2h44il2eivuy

Deep Learning for Time Series Forecasting: Tutorial and Literature Survey

Konstantinos Benidis, Syama Sundar Rangapuram, Valentin Flunkert, Yuyang Wang, Danielle Maddix, Caner Turkmen, Jan Gasthaus, Michael Bohlke-Schneider, David Salinas, Lorenzo Stella, François-Xavier Aubet, Laurent Callot (+1 others)
2022 ACM Computing Surveys  
Consequently, over the last years, these methods are now ubiquitous in large-scale industrial forecasting applications and have consistently ranked among the best entries in forecasting competitions (e.g  ...  In this article we provide an introduction and overview of the field: We present important building blocks for deep forecasting in some depth; using these building blocks, we then survey the breadth of  ...  In an early example of hierarchical Bayesian models, [31] combined global and local features for intermittent demand forecasting in retail planning.  ... 
doi:10.1145/3533382 fatcat:l46f34dbp5fdpawbpnoiippf6q

Retail forecasting: Research and practice

Robert Fildes, Shaohui Ma, Stephan Kolassa
2019 International Journal of Forecasting  
Forecasting with temporally aggregated demand signals in a retail supply chain.  ...  sales and potential demand.  ...  latent state model of on-line demand for Amazon products.  ... 
doi:10.1016/j.ijforecast.2019.06.004 fatcat:qrrydkizzjg3jmuq7tnpxtlloa

Bayesian forecasting of parts demand

Phillip M. Yelland
2010 International Journal of Forecasting  
The paper describes a Bayesian statistical model developed to forecast parts demand for Sun Microsystems, Inc., a major vendor of network computer products.  ...  Furthermore, using hierarchical priors, the model is able to pool demand patterns for a collection of parts, producing calibrated forecasts for new parts with little or no demand history.  ...  , such as ARIMA 4 (Gilbert 2005) or linear/Gaussian state-space (Aviv 2003).  ... 
doi:10.1016/j.ijforecast.2009.11.001 fatcat:enlopuhpm5h5pknxoylmindrnu

A Spliced Gamma-Generalized Pareto Model for Short-Term Extreme Wind Speed Probabilistic Forecasting

Daniela Castro-Camilo, Raphaël Huser, Håvard Rue
2019 Journal of Agricultural Biological and Environmental Statistics  
Our model belongs to the class of latent Gaussian models, for which inference is conveniently performed based on the integrated nested Laplace approximation method.  ...  In this work, we develop a flexible spliced Gamma-Generalized Pareto model to forecast extreme and non-extreme wind speeds simultaneously.  ...  ACKNOWLEDGEMENTS We thank Amanda Hering for helpful suggestions, and for providing the wind speed data. We also extend our thanks to Thomas Opitz for helpful discussion.  ... 
doi:10.1007/s13253-019-00369-z fatcat:cjyeo4my7zes3cqlbat7xbhhwe

A Bayesian processor of uncertainty for precipitation forecasting using multiple predictors and censoring

Paolo Reggiani, Oleksiy Boyko
2019 Monthly Weather Review  
A Bayesian processor of uncertainty for numerical precipitation forecasts is presented.  ...  Standard forecast performance evaluation and verification metrics are employed to set the approach into perspective against Bayesian model averaging (BMA).  ...  We also acknowledge Meteo Swiss and ECMWF for giving access to the data used in this study.  ... 
doi:10.1175/mwr-d-19-0066.1 fatcat:potufbksz5g4rdcnet4chscuky
« Previous Showing results 1 — 15 out of 776 results