DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks

2017/04/13 by Salinas, David, Flunkert, Valentin, Gasthaus, Jan · 114 citations
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1704.04110

Abstract

Probabilistic forecasting, i.e. estimating the probability distribution of a time series' future given its past, is a key enabler for optimizing business processes. In retail businesses, for example, forecasting demand is crucial for having the right inventory available at the right time at the right place. In this paper we propose DeepAR, a methodology for producing accurate probabilistic forecasts, based on training an auto regressive recurrent network model on a large number of related time series. We demonstrate how by applying deep learning techniques to forecasting, one can overcome many of the challenges faced by widely-used classical approaches to the problem. We show through extensive empirical evaluation on several real-world forecasting data sets accuracy improvements of around 15% compared to state-of-the-art methods.

Cited by

• What Matters in Deep Learning for Time Series Forecasting?
• QuantFlow: A Federated Mamba-Based Post-Transformer Foundation Model for Time-Series Forecasting
• Foundation Models and Fine-Tuning: Toward a New Generation of Models for Time Series Forecasting
• Deep Generative Spatiotemporal Engression for Probabilistic Forecasting of Epidemics
• RIPCN: A Road Impedance Principal Component Network for Probabilistic Traffic Flow Forecasting
• Dynamic Attention (DynAttn): Interpretable High-Dimensional Spatio-Temporal Forecasting (with Application to Conflict Fatalities)
• DecoKAN: Interpretable Decomposition for Forecasting Cryptocurrency Market Dynamics
• LoFT-LLM: Low-Frequency Time-Series Forecasting with Large Language Models
• Quantitative mobile gamma-ray spectrometry through Bayesian inference
• An End-to-End Approach for Microgrid Probabilistic Forecasting and Robust Operation via Decision-focused Learning
• HydroDiffusion: Diffusion-Based Probabilistic Streamflow Forecasting with a State Space Backbone
• Mitigating Exposure Bias in Risk-Aware Time Series Forecasting with Soft Tokens
• Identifying Neutron Sources using Recoil and Time-of-Flight Spectroscopy
• FRWKV:Frequency-Domain Linear Attention for Long-Term Time Series Forecasting
• A Comparative Study on How Data Normalization Affects Zero-Shot Generalization in Time Series Foundation Models
• A Multi-View Multi-Timescale Hypergraph-Empowered Spatiotemporal Framework for EV Charging Forecasting
• Robust Probabilistic Load Forecasting for a Single Household: A Comparative Study from SARIMA to Transformers on the REFIT Dataset
• TARFVAE: Efficient One-Step Generative Time Series Forecasting via TARFLOW based VAE
• Time Series Forecasting via Direct Per-Step Probability Distribution Modeling
• Modèles de Fondation et Ajustement : Vers une Nouvelle Génération de Modèles pour la Prévision des Séries Temporelles
• TS2Vec-Ensemble: An Enhanced Self-Supervised Framework for Time Series Forecasting
• SimDiff: Simpler Yet Better Diffusion Model for Time Series Point Forecasting
• WaveTuner: Comprehensive Wavelet Subband Tuning for Time Series Forecasting
• Structured Noise Modeling for Enhanced Time-Series Forecasting
• TimePre: Bridging Accuracy, Efficiency, and Stability in Probabilistic Time-Series Forecasting
• PeriodNet: Boosting the Potential of Attention Mechanism for Time Series Forecasting
• Multi-layer Stack Ensembles for Time Series Forecasting
• Stochastic Predictive Analytics for Stocks in the Newsvendor Problem
• FreDN: Spectral Disentanglement for Time Series Forecasting via Learnable Frequency Decomposition
• VitalBench: A Rigorous Multi-Center Benchmark for Long-Term Vital Sign Prediction in Intraoperative Care
• Moirai 2.0: When Less Is More for Time Series Forecasting
• LAD-BNet: Lag-Aware Dual-Branch Networks for Real-Time Energy Forecasting on Edge Devices
• DeepBooTS: Dual-Stream Residual Boosting for Drift-Resilient Time-Series Forecasting
• Predicting the Future by Retrieving the Past
• Synapse: Adaptive Arbitration of Complementary Expertise in Time Series Foundational Models
• Agentic World Modeling for 6G: Near-Real-Time Generative State-Space Reasoning
• Contextual Deconvolution for Variance-Stable Demand Sensing: Kernel-Modulated Operators in Promotional Retail
• ProbFM: Probabilistic Time Series Foundation Model with Uncertainty Decomposition
• ARIMA_PLUS: Large-scale, Accurate, Automatic and Interpretable In-Database Time Series Forecasting and Anomaly Detection in Google BigQuery
• Group Interventions on Deep Networks for Causal Discovery in Subsystems
• NetBurst: Event-Centric Forecasting of Bursty, Intermittent Time Series
• Chronos-2: From Univariate to Universal Forecasting
• SynTSBench: Rethinking Temporal Pattern Learning in Deep Learning Models for Time Series
• Speculative Sampling for Parametric Temporal Point Processes
• SEMPO: Lightweight Foundation Models for Time Series Forecasting
• Foundation Model Forecasts: Form and Function
• QKCV Attention: Enhancing Time Series Forecasting with Static Categorical Embeddings for Both Lightweight and Pre-trained Foundation Models
• Online Time Series Forecasting with Theoretical Guarantees
• Benchmarking Probabilistic Time Series Forecasting Models on Neural Activity
• Xihe: Scalable Zero-Shot Time Series Learner Via Hierarchical Interleaved Block Attention
• Tackling Time-Series Forecasting Generalization via Mitigating Concept Drift
• Rethinking Evaluation in the Era of Time Series Foundation Models: (Un)known Information Leakage Challenges
• CoRA: Covariate-Aware Adaptation of Time Series Foundation Models
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting
• STaTS: Structure-Aware Temporal Sequence Summarization via Statistical Window Merging
• UAMDP: Uncertainty-Aware Markov Decision Process for Risk-Constrained Reinforcement Learning from Probabilistic Forecasts
• Inner-Instance Normalization for Time Series Forecasting
• HTMformer: Hybrid Time and Multivariate Transformer for Time Series Forecasting
• CNN-TFT explained by SHAP with multi-head attention weights for time series forecasting
• Test-Time Efficient Pretrained Model Portfolios for Time Series Forecasting
• Smart Contract Adoption under Discrete Overdispersed Demand: A Negative Binomial Optimization Perspective
• ResCP: Reservoir Conformal Prediction for Time Series Forecasting
• Forecasting-based Biomedical Time-series Data Synthesis for Open Data and Robust AI
• Forking-Sequences
• TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis
• SSM-CGM: Interpretable State-Space Forecasting Model of Continuous Glucose Monitoring for Personalized Diabetes Management
• From News to Returns: A Granger-Causal Hypergraph Transformer on the Sphere
• Predictive inference for time series: why is split conformal effective despite temporal dependence?
• Understanding Transformers for Time Series: Rank Structure, Flow-of-ranks, and Compressibility
• Integrating AI and Ensemble Forecasting: Explainable Materials Planning with Scorecards and Trend Insights for a Large-Scale Manufacturer
• fev-bench: A Realistic Benchmark for Time Series Forecasting
• ReNF: Rethinking the Design of Neural Long-Term Time Series Forecasters
• DSAT-HD: Dual-Stream Adaptive Transformer with Hybrid Decomposition for Multivariate Time Series Forecasting
• From Noise to Laws: Regularized Time-Series Forecasting via Denoised Dynamic Graphs
• Dynamic Lagging for Time-Series Forecasting in E-Commerce Finance: Mitigating Information Loss with A Hybrid ML Architecture
• Pi-Transformer: A Physics-informed Attention Mechanism for Time Series Anomaly Detection
• From Samples to Scenarios: A New Paradigm for Probabilistic Forecasting
• Latent Matters: Learning Deep State-Space Models
• A More Realistic Evaluation of Cross-Frequency Transfer Learning and Foundation Forecasting Models
• An AutoML Framework using AutoGluonTS for Forecasting Seasonal Extreme Temperatures
• Tsururu: A Python-based Time Series Forecasting Strategies Library
• ScenGAN: Attention-Intensive Generative Model for Uncertainty-Aware Renewable Scenario Forecasting
• Super-Linear: A Lightweight Pretrained Mixture of Linear Experts for Time Series Forecasting
• TFMAdapter: Lightweight Instance-Level Adaptation of Foundation Models for Forecasting with Covariates
• A Time-Series Foundation Model by Universal Delay Embedding
• GFS: A Preemption-aware Scheduling Framework for GPU Clusters with Predictive Spot Instance Management
• Towards Scalable and Structured Spatiotemporal Forecasting
• Neural ARFIMA model for forecasting BRIC exchange rates with long memory
• ARIES: Relation Assessment and Model Recommendation for Deep Time Series Forecasting
• VARMA-Enhanced Transformer for Time Series Forecasting
• MillGNN: Learning Multi-Scale Lead-Lag Dependencies for Multi-Variate Time Series Forecasting
• Conformal Prediction for Time-series Forecasting with Change Points
• StoxLSTM: A Stochastic Extended Long Short-Term Memory Network for Time Series Forecasting
• Forecasting Probability Distributions of Financial Returns with Deep Neural Networks
• Frozen in Time: Parameter-Efficient Time Series Transformers via Reservoir-Induced Feature Expansion and Fixed Random Dynamics
• Investment Portfolio Optimization Based on Modern Portfolio Theory and Deep Learning Models
• Hierarchical Evaluation Function: A Multi-Metric Approach for Optimizing Demand Forecasting Models
• Probabilistic Forecasting Method for Offshore Wind Farm Cluster under Typhoon Conditions: a Score-Based Conditional Diffusion Model
• TimeMKG: Knowledge-Infused Causal Reasoning for Multivariate Time Series Modeling
• Measuring Time Series Forecast Stability for Demand Planning
• Adaptive Fine-Tuning via Pattern Specialization for Deep Time Series Forecasting
• Comparative Analysis of Time Series Foundation Models for Demographic Forecasting: Enhancing Predictive Accuracy in US Population Dynamics
• From Values to Tokens: An LLM-Driven Framework for Context-aware Time Series Forecasting via Symbolic Discretization
• EnergyPatchTST: Multi-scale Time Series Transformers with Uncertainty Estimation for Energy Forecasting
• VisionTS++: Cross-Modal Time Series Foundation Model with Continual Pre-trained Vision Backbones
• STARE: Predicting Decision Making Based on Spatio-Temporal Eye Movements
• Quantum Temporal Fusion Transformer
• Supervised Dynamic Dimension Reduction with Deep Neural Network
• DeepKoopFormer: A Koopman Enhanced Transformer Based Architecture for Time Series Forecasting
• Signals, Concepts, and Laws: Toward Universal, Explainable Time-Series Forecasting
• Foundation Models for Demand Forecasting via Dual-Strategy Ensembling
• Spatiodynamic inference using vision-based generative modelling
• Adaptive Fuzzy Time Series Forecasting via Partially Asymmetric Convolution and Sub-Sliding Window Fusion
• SPADE-S: A Sparsity-Robust Foundational Forecaster

Related

• DeepAR: Probabilistic forecasting with autoregressive recurrent networks
• Forecasting with Deep Learning
• DANLIP: Deep Autoregressive Networks for Locally Interpretable Probabilistic Forecasting
• Learning Interpretable Deep State Space Model for Probabilistic Time Series Forecasting
• Deep learning with long short-term memory networks and random forests for demand forecasting in multi-channel retail