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

By Matthias W. Seeger, Syama Sundar Rangapuram, Bernie Wang, David Salinas, Jan Gasthaus, Tim Januschowski, Valentin Flunkert. 2017

20 citations recorded by the source corpus.

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. 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. In a study on large real-world sales datasets, our method outperforms competing approaches on fast and medium moving items.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep Factors for Forecasting (2019, methodology): “Our algorithms are implemented in MXNet Gluon (Chen et al., 2015), and make extensive use of its Linear Algebra library (Seeger et al., 2017; Zhenwen et al., 2018).”
Forecasting Big Time Series: Old and New (2018, background): “In particular, we focus on the exponential smoothing models (Innovation state space models (ISSM)) [11] and structural time series models [8, 5, 30] along with their Bayesian counterparts [32, 31].”
Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting (2020, methodology): “ing (Box et al. 2015; Ray 1990; Seeger et al. 2017; Seeger, Salinas, and Flunkert 2016), and deep learning techniques mainly develop an encoder-decoder prediction paradigm by using RNN and their variants (Hochreiter and Schmidhuber 1997; Li et al.”
Deep State Space Models for Time Series Forecasting (2018, methodology): “We encode the day-of-week seasonality using seven components of the latent state as in [22], i.”