README.md

List of state of the art papers, code, and other resources focus on time series forecasting.

Table of Contents

• M4 competition
• Kaggle time series competition
• Papers
• Conferences
• Theory-Resource
• Code Resource
• Datasets

M4-competition

M4

papers

• The M4 Competition: 100,000 time series and 61 forecasting methods
• A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting
• Weighted ensemble of statistical models
• FFORMA: Feature-based forecast model averaging

Kaggle-time-series-competition

• Walmart Store Sales Forecasting (2014)
• Walmart Sales in Stormy Weather (2015)
• Rossmann Store Sales (2015)
• Wikipedia Web Traffic Forecasting (2017)
• Corporación Favorita Grocery Sales Forecasting (2018)
• Recruit Restaurant Visitor Forecasting (2018)
• COVID19 Global Forecasting (2020)
• Jane Street Future Market Prediction(2021)

Papers

2024

• TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting Arxiv

  • [code]

• UniTS: Building a Unified Time Series Model Arxiv

  • [code]

• Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting Arxiv

  • [code]

2023

• MICN: Multi-scale Local and Global Context Modeling for Long-term Series Forecasting ICLR 2023 Oral

  • [code]

• Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting ICLR 2023

  • [code]

• Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecasting ICLR 2023

  • [code]

• SAITS: Self-Attention-based Imputation for Time Series Expert Systems with Applications

  • [code]

• A Time Series is Worth 64 Words: Long-term Forecasting with Transformers ICLR 2023

  • [code]

2022

• Deep Learning for Time Series Anomaly Detection: A Survey survey

• A Comprehensive Survey of Regression Based Loss Functions for Time Series Forecasting survey

  • [code]

• Non-stationary Transformers: Rethinking the Stationarity in Time Series Forecasting NeurIPS 2022

• Generative Time Series Forecasting with Diffusion, Denoise and Disentanglement NeurIPS 2022

• SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction NeurIPS 2022

• Learning Latent Seasonal-Trend Representations for Time Series Forecasting NeurIPS 2022

• GT-GAN: General Purpose Time Series Synthesis with Generative Adversarial Networks NeurIPS 2022

• Causal Disentanglement for Time Series NeurIPS 2022

• Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency NeurIPS 2022

• FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecasting NeurIPS 2022

• BILCO: An Efficient Algorithm for Joint Alignment of Time Series NeurIPS 2022

• LSAR: Efficient Leverage Score Sampling Algorithm for the Analysis of Big Time Series Data NeurIPS 2022

• Unsupervised Learning of Algebraic Structure from Stationary Time Sequences NeurIPS 2022

• Dynamic Sparse Network for Time Series Classification: Learning What to “See” NeurIPS 2022

• WaveBound: Dynamically Bounding Error for Stable Time Series Forecasting NeurIPS 2022

• Conditional Loss and Deep Euler Scheme for Time Series Generation AAAI 2022

• I-SEA: Importance Sampling and Expected Alignment-Based Deep Distance Metric Learning for Time Series Analysis and Embedding AAAI 2022

• TS2Vec: Towards Universal Representation of Time Series AAAI 2022

• Reinforcement Learning based Dynamic Model Combination for Time Series Forecasting AAAI 2022

• CATN: Cross Attentive Tree-Aware Network for Multivariate Time Series Forecasting AAAI 2022

• Transformers in Time Series: A Survey review

  • Wen, et al.
  • Code

• Pyraformer: Low-Complexity Pyramidal Attention for Long-Range Time Series Modeling and Forecasting ICLR 2022 oral

  • Liu, et al.

2021

• A machine learning approach for forecasting hierarchical time series

  • Mancuso, et al.

• Probabilistic Transformer For Time Series Analysis NeuIPS 2021

  • Tang, et al.

• Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting NeuIPS 2021

  • Wu, et al.

• CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation NeuIPS 2021

  • Yusuke, et al.

• Variational Inference for Continuous-Time Switching Dynamical Systems NeuIPS 2021

  • Lukas, et al.

• MixSeq: Connecting Macroscopic Time Series Forecasting with Microscopic Time Series Data NeuIPS 2021

  • Zhu, et al.

• Coresets for Time Series Clustering NeuIPS 2021

  • Zhou, et al.

• Online false discovery rate control for anomaly detection in time series NeuIPS 2021

  • Quentin, et al.

• Adjusting for Autocorrelated Errors in Neural Networks for Time Series NeuIPS 2021

  • Sun, et al.

• Deep Explicit Duration Switching Models for Time Series NeuIPS 2021

  • Zhou, et al.

• Deep Learning for Time Series Forecasting: A Survey survey

  • Torres, et al.

• Whittle Networks: A Deep Likelihood Model for Time Series ICML 2021

  • Yu, et al.
  • Code

• Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting ICML 2021

  • Chen, et al.
  • Code

• Long Horizon Forecasting With Temporal Point Processes WSDM 2021

  • Deshpande, et al.
  • Code

• Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting AAAI 2021 best paper

  • Zhou, et al.
  • Code

• Coupled Layer-wise Graph Convolution for Transportation Demand Prediction AAAI 2021

  • Ye, et al.
  • Code

2020

• Block Hankel Tensor ARIMA for Multiple Short Time Series Forecasting AAAI 2020

  • Shi, et al.
  • Code

• Adversarial Sparse Transformer for Time Series Forecasting NeurIPS 2020

  • Wu, et al.
  • Code not yet

• Benchmarking Deep Learning Interpretability in Time Series Predictions NeurIPS 2020

  • Ismail, et al.
  • [Code]

• Deep reconstruction of strange attractors from time series NeurIPS 2020

  • Gilpin, et al.
  • [Code]

• Rethinking 1D-CNN for Time Series Classification: A Stronger Baseline classification

  • Tang, et al.
  • [Code]

• Active Model Selection for Positive Unlabeled Time Series Classification

  • Liang, et al.
  • [Code]

• Unsupervised Phase Learning and Extraction from Quasiperiodic Multidimensional Time-series Data

  • Prayook, et al.
  • [Code]

• Connecting the Dots: Multivariate Time Series Forecasting withGraph Neural Networks

  • Wu, et al.
  • [Code]

• Forecasting with sktime: Designing sktime's New Forecasting API and Applying It to Replicate and Extend the M4 Study

  • Löning, et al.
  • Code not yet

• RobustTAD: Robust Time Series Anomaly Detection viaDecomposition and Convolutional Neural Networks

  • Gao, et al.
  • Code not yet

• Neural Controlled Differential Equations forIrregular Time Series

  • Patrick Kidger, et al.
  • University of Oxford
  • [Code]

• Time Series Forecasting With Deep Learning: A Survey

  • Lim, et al.
  • Code not yet

• Neural forecasting: Introduction and literature overview

  • Benidis, et al.
  • Amazon Research
  • Code not yet.

• Time Series Data Augmentation for Deep Learning: A Survey

  • Wen, et al.
  • Code not yet

• Modeling time series when some observations are zeroJournal of Econometrics 2020

  • Andrew Harveyand Ryoko Ito.
  • Code not yet

• Meta-learning framework with applications to zero-shot time-series forecasting

  • Oreshkin, et al.
  • Code not yet.

• Harmonic Recurrent Process for Time Series Forecasting

  • Shao-Qun Zhang and Zhi-Hua Zhou.
  • Code not yet.

• Block Hankel Tensor ARIMA for Multiple Short Time Series ForecastingAAAI 2020

  • QIQUAN SHI, et al.
  • Code not yet

• Learnings from Kaggle's Forecasting Competitions

  • Casper Solheim Bojer, et al.
  • Code not yet.

• An Industry Case of Large-Scale Demand Forecasting of Hierarchical Components

  • Rodrigo Rivera-Castro, et al.
  • Code not yet.

• Multi-variate Probabilistic Time Series Forecasting via Conditioned Normalizing Flows

  • Kashif Rasul, et al.
  • Code not yet.

• ForecastNet: A Time-Variant Deep Feed-Forward Neural Network Architecture for Multi-Step-Ahead Time-Series Forecasting

  • Joel Janek Dabrowski, et al.
  • Code not yet.

• Anomaly detection for Cybersecurity: time series forecasting and deep learningGood review about forecasting

  • Giordano Colò.
  • Code not yet.

• Event-Driven Continuous Time Bayesian Networks

  • Debarun Bhattacharjya, et al.
  • Research AI, IBM
  • Code not yet.

Conferences

• ICLR
• AAAI
• IJCAI
• ISF
• NeurIPS
• ICML
• M5 Competition

Theory-Resource

• Time Series Analysis, MIT

• Time Series Forecasting, Udacity

• Practical Time Series Analysis, Cousera

• Sequences, Time Series and Prediction

• Intro to Time Series Analysis in R, Cousera

• Anomaly Detection in Time Series Data with Keras, Corsera

• Applying Data Analytics in Finance, Coursera

• Time Series Forecasting using Python

• STAT 510: Applied Time Series Analysis, PSU

• Policy Analysis Using Interrupted Time Series, edx

• Time Series Forecasting in Python

• time-series-transformers-review

Code-Resource

• PyPOTS: A Python Toolbox for Data Mining on Partially-Observed Time Series

• FOST from microsoft

• pyWATTS: Python Workflow Automation Tool for Time-Series

• Seglearn: A Python Package for Learning Sequences and Time Series

• tsflex: Flexible Time Series Processing & Feature Extraction

• cesium: Open-Source Platform for Time Series Inference

• PyTorch Forecasting: A Python Package for time series forecasting with PyTorch

• A collection of time series prediction methods: rnn, seq2seq, cnn, wavenet, transformer, unet, n-beats, gan, kalman-filter

• Implementation of Transformer model (originally from Attention is All You Need) applied to Time Series

• Predicting/hypothesizing the findings of the M4 Competition

• PyFlux

• HyperTS: A Full-Pipeline Automated Time Series Analysis Toolkit

• Time Series Forecasting Best Practices & Examples

• List of tools & datasets for anomaly detection on time-series data

• python packages for time series analysis

• A scikit-learn compatible Python toolbox for machine learning with time series

• plotly-resampler: Visualize large time series data with plotly.py

• time series visualization tools

• A statistical library designed to fill the void in Python's time series analysis capabilities

• RNN based Time-series Anomaly detector model implemented in Pytorch

• ARCH models in Python

• A Python toolkit for rule-based/unsupervised anomaly detection in time series

• A curated list of awesome time series databases, benchmarks and papers

• Matrix Profile analysis methods in Python for clustering, pattern mining, and anomaly detection

• Flow Forecast: A deep learning framework for time series forecasting, classification and anomaly detection built in PyTorch

Datasets

• TSDB: A Python Toolbox to Ease Loading Open-Source Time-Series Datasets (supporting 119 datasets)

• SkyCam: A Dataset of Sky Images and their Irradiance values

• U.S. Air Pollution Data

• U.S. Chronic Disease Data

• Air quality from UCI

• Seattle freeway traffic speed

• Youth Tobacco Survey Data

• Singapore Population

• Airlines Delay

• Airplane Crashes

• Electricity dataset from UCI

• Traffic dataset from UCI

• City of Baltimore Crime Data

• Discover The Menu

• Global Climate Change Data

• Global Health Nutrition Data

• Beijing PM2.5 Data Set

• Airline Passengers dataset

• Government Finance Statistics

• Historical Public Debt Data

• Kansas City Crime Data

• NYC Crime Data

• Kaggle-Web Traffic Time Series Forecasting