mlsanityData to 7 Oct 2026

Coherent Probabilistic Aggregate Queries on Long-horizon Forecasts

arXiv 2111.03394 ↗First posted 5 Nov 2021cs.LG stat.MLcited by 0

Not on the map. Why?

Abstract (arXiv)

Long range forecasts are the starting point of many decision support systems that need to draw inference from high-level aggregate patterns on forecasted values. State of the art time-series forecasting methods are either subject to concept drift on long-horizon forecasts, or fail to accurately predict coherent and accurate high-level aggregates. In this work, we present a novel probabilistic forecasting method that produces forecasts that are coherent in terms of base level and predicted aggregate statistics. We achieve the coherency between predicted base-level and aggregate statistics using a novel inference method based on KL-divergence that can be solved efficiently in closed form. We show that our method improves forecast performance across both base level and unseen aggregates post inference on real datasets ranging three diverse domains. (hrefhttps://github.com/pratham16cse/AggForecasterProject URL)

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 15 of its 23 references that are in the map. First posted on arXiv 5 Nov 2021.

Builds on

15 of 23 references.

References · 15

  1. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  2. 2017DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networkscited by 1,004
  3. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
  4. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 744
  5. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 606
  6. 2017A Multi-Horizon Quantile Recurrent Forecastercited by 211
  7. 2019High-Dimensional Multivariate Forecasting with Low-Rank Gaussian Copula Processescited by 147
  8. 2019Shape and Time Distortion Loss for Training Deep Time Series Forecasting Modelscited by 66
  9. 2021Missing Value Imputation on Multidimensional Time Seriescited by 33
  10. 2021Simultaneously Reconciled Quantile Forecasting of Hierarchically Related Time Seriescited by 24
  11. 2021Long Horizon Forecasting With Temporal Point Processescited by 20
  12. 2017The Statistical Recurrent Unitcited by 10
3 more
  1. 2018ARMDN: Associative and Recurrent Mixture Density Networks for eRetail Demand Forecastingcited by 9
  2. 2020Probabilistic Time Series Forecasting with Structured Shape and Temporal Diversitycited by 9
  3. 2019Streaming Adaptation of Deep Forecasting Models using Adaptive Recurrent Unitscited by 5

Built on by

Not cited yet.