Awesome Time Series Forecasting/Prediction Papers
This repository contains a reading list of papers (285 papers !!! ) on Time Series Forecasting/Prediction (TSF) and Spatio-Temporal Forecasting/Prediction (STF) . These papers are mainly categorized according to the type of model. This repository is still being continuously improved. In addition to papers that have been accepted by top conferences or journals, the repository also includes the latest papers from arXiv . If you have found any relevant papers that need to be included in this repository, please feel free to submit a pull request (PR) or open an issue. If you find this repository useful, please give it a π.
Each paper may apply to one or several types of forecasting, including univariate time series forecasting, multivariate time series forecasting, and spatio-temporal forecasting, which are also marked in the Type column. If covariates and exogenous variables are not considered , univariate time series forecasting involves predicting the future of one variable with the history of this variable, while multivariate time series forecasting involves predicting the future of C variables with the history of C variables. Note that repeating univariate forecasting multiple times can also achieve the goal of multivariate forecasting, which is called channel-independent . However, univariate forecasting methods cannot extract relationships between variables, so the basis for distinguishing between univariate and multivariate forecasting methods is whether the method involves interaction between variables. Besides, in the era of deep learning, many univariate models can be easily modified to directly process multiple variables for multivariate forecasting. And multivariate models generally can be directly used for univariate forecasting. Here we classify solely based on the model's description in the original paper. Spatio-temporal forecasting is often used in traffic and weather forecasting, and it adds a spatial dimension compared to univariate and multivariate forecasting. In spatio-temporal forecasting, if each measurement point has only one variable, it is equivalent to multivariate forecasting. Therefore, the distinction between spatio-temporal forecasting and multivariate forecasting is not clear. Spatio-temporal models can usually be directly applied to multivariate forecasting, and multivariate models can also be used for spatio-temporal forecasting with minor modifications. Here we also classify solely based on the model's description in the original paper.
univariate time series forecasting: , where L is the history length, H is the prediction horizon length.
multivariate time series forecasting: , where C is the number of variables (channels).
spatio-temporal forecasting: , where N is the spatial dimension (number of measurement points).
irregular time series: observation/sampling times are irregular.
Some Additional Information.
π© 2023/11/1: I have marked some recommended papers with π (Just my personal preference π).
π© 2023/11/1: I have added a new category : models specifically designed for irregular time series.
π© 2023/11/1: I also recommend you to check out some other GitHub repositories about awesome time series papers: time-series-transformers-review , awesome-AI-for-time-series-papers , time-series-papers , deep-learning-time-series .
π© 2023/11/3: There are some popular toolkits or code libraries that integrate many time series models: Time-Series-Library , Prophet , Darts , Kats , tsai , GluonTS , PyTorchForecasting , tslearn , AutoGluon , flow-forecast , PyFlux .
π© 2023/12/28: Since the topic of LLM(Large Language Model)+TS(Time Series) has been popular recently, I have introduced a category (LLM) to include related papers. This is distinguished from the Pretrain category. Pretrain mainly contains papers which design agent tasks (contrastive or generative) suitable for time series, and only use large-scale time series data for pre-training.
π© 2024/4/1: Some researchers have introduced the recently popular Mamba model into the field of time series forecasting, which can be found in the SSM (State Space Model) table.
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
15-11-23
Multi-step
ACOMP 2015
Comparison of Strategies for Multi-step-Ahead Prediction of Time Series Using Neural Network
None
19-06-20
DL
SENSJ 2019
A Review of Deep Learning Models for Time Series Prediction
None
20-09-27
DL
Arxiv 2020
Time Series Forecasting With Deep Learning: A Survey
None
22-02-15
Transformer
IJCAI 2023
Transformers in Time Series: A Survey
PaperList
23-03-25
STGNN
Arxiv 2023
Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Survey
None
23-05-01
Diffusion
Arxiv 2023
Diffusion Models for Time Series Applications: A Survey
None
23-06-16
SSL
TPAMI 2024
Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects
None
23-06-20
OpenSTL
NIPS 2023
OpenSTL: A Comprehensive Benchmark of Spatio-Temporal Predictive Learning
Benchmark
23-07-07
GNN
Arxiv 2023
A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detection
PaperList
23-10-09
BasicTS
Arxiv 2023
Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysis
Benchmark
23-10-11
ProbTS
Arxiv 2023
ProbTS: A Unified Toolkit to Probe Deep Time-series Forecasting
Toolkit
23-12-28
TSPP
Arxiv 2023
TSPP: A Unified Benchmarking Tool for Time-series Forecasting
TSPP
24-01-05
Diffusion
Arxiv 2024
The Rise of Diffusion Models in Time-Series Forecasting
None
24-02-15
LLM
Arxiv 2024
Large Language Models for Forecasting and Anomaly Detection: A Systematic Literature Review
None
24-03-21
FM
Arxiv 2024
Foundation Models for Time Series Analysis: A Tutorial and Survey
None
24-03-29
TFB
VLDB 2024
TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methods
TFB
24-04-24
Mamba-360
Arxiv 2024
Mamba-360: Survey of State Space Models as Transformer Alternative for Long Sequence Modelling: Methods, Applications, and Challenges
Mamba-360
24-04-29
Diffusion
Arxiv 2024
A Survey on Diffusion Models for Time Series and Spatio-Temporal Data
PaperList
Date
Method
Type
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
17-09-14
STGCN π
IJCAI 2018
Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting
STGCN
19-05-31
Graph WaveNet
IJCAI 2019
Graph WaveNet for Deep Spatial-Temporal Graph Modeling
Graph-WaveNet
19-07-17
ASTGCN
AAAI 2019
Attention Based Spatial-Temporal Graph Convolutional Networks for Traffic Flow Forecasting
ASTGCN
20-04-03
SLCNN
AAAI 2020
Spatio-Temporal Graph Structure Learning for Traffic Forecasting
None
20-04-03
GMAN
AAAI 2020
GMAN: A Graph Multi-Attention Network for Traffic Prediction
GMAN
20-05-03
MTGNN π
KDD 2020
Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks
MTGNN
21-03-13
StemGNN π
NIPS 2020
Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting
StemGNN
22-05-16
TPGNN
NIPS 2022
Multivariate Time-Series Forecasting with Temporal Polynomial Graph Neural Networks
TPGNN
22-06-18
D2STGNN
VLDB 2022
Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting
D2STGNN
23-05-12
DDGCRN
PR 2023
A Decomposition Dynamic graph convolutional recurrent network for traffic forecasting
DDGCRN
23-07-10
NexuSQN
Arxiv 2023
Nexus sine qua non: Essentially connected neural networks for spatial-temporal forecasting of multivariate time series
None
23-11-10
FourierGNN
NIPS 2023
FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspective
FourierGNN
23-12-05
SAMSGL
TETCI 2023
SAMSGL: Series-Aligned Multi-Scale Graph Learning for Spatio-Temporal Forecasting
None
23-12-27
TGCRN
ICDE 2024
Learning Time-aware Graph Structures for Spatially Correlated Time Series Forecasting
None
23-12-27
FCDNet
Arxiv 2023
FCDNet: Frequency-Guided Complementary Dependency Modeling for Multivariate Time-Series Forecasting
FCDNet
23-12-31
MSGNet
AAAI 2024
MSGNet: Learning Multi-Scale Inter-Series Correlations for Multivariate Time Series Forecasting
MSGNet
24-01-15
RGDAN
NN 2024
RGDAN: A random graph diffusion attention network for traffic prediction
RGDAN
24-01-16
BiTGraph
ICLR 2024
Biased Temporal Convolution Graph Network for Time Series Forecasting with Missing Values
BiTGraph
24-01-24
TMP
AAAI 2024
Time-Aware Knowledge Representations of Dynamic Objects with Multidimensional Persistence
None
24-02-16
HD-TTS
Arxiv 2024
Graph-based Forecasting with Missing Data through Spatiotemporal Downsampling
None
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
18-05-18
DSSM
NIPS 2018
Deep State Space Models for Time Series Forecasting
None
19-08-10
DSSMF
IJCAI 2019
Learning Interpretable Deep State Space Model for Probabilistic Time Series Forecasting
None
22-08-19
SSSD
TMLR 2022
Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models
SSSD
22-09-22
SpaceTime
ICLR 2023
Effectively Modeling Time Series with Simple Discrete State Spaces
SpaceTime
22-12-24
LS4
ICML 2023
Deep Latent State Space Models for Time-Series Generation
LS4
24-02-18
Attraos
Arxiv 2024
Attractor Memory for Long-Term Time Series Forecasting: A Chaos Perspective
None
24-03-14
TimeMachine
Arxiv 2024
TimeMachine: A Time Series is Worth 4 Mambas for Long-term Forecasting
None
24-03-17
S-D-Mamba
Arxiv 2024
Is Mamba Effective for Time Series Forecasting?
S-D-Mamba
24-03-29
TSM2
Arxiv 2024
MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selection
M2
24-04-23
Mambaformer
Arxiv 2024
Integrating Mamba and Transformer for Long-Short Range Time Series Forecasting
Mambaformer
24-04-24
Bi-Mamba4TS
Arxiv 2024
Bi-Mamba4TS: Bidirectional Mamba for Time Series Forecasting
None
Plug and Play (Model-Agnostic).
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
19-02-21
DAIN π
TNNLS 2020
Deep Adaptive Input Normalization for Time Series Forecasting
DAIN
19-09-19
DILATE
NIPS 2019
Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models
DILATE
21-07-19
TAN
NIPS 2021
Topological Attention for Time Series Forecasting
TAN
21-09-29
RevIN π
ICLR 2022
Reversible Instance Normalization for Accurate Time-Series Forecasting against Distribution Shift
RevIN
22-02-23
MQF2
AISTATS 2022
Multivariate Quantile Function Forecaster
None
22-05-18
FiLM
NIPS 2022
FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecasting
FiLM
23-02-18
FrAug
Arxiv 2023
FrAug: Frequency Domain Augmentation for Time Series Forecasting
FrAug
23-02-22
Dish-TS
AAAI 2023
Dish-TS: A General Paradigm for Alleviating Distribution Shift in Time Series Forecasting
Dish-TS
23-02-23
Adaptive Sampling
NIPSW 2022
Adaptive Sampling for Probabilistic Forecasting under Distribution Shift
None
23-04-19
RoR
ICML 2023
Regions of Reliability in the Evaluation of Multivariate Probabilistic Forecasts
RoR
23-05-26
BetterBatch
Arxiv 2023
Better Batch for Deep Probabilistic Time Series Forecasting
None
23-05-28
PALS
Arxiv 2023
Adaptive Sparsity Level during Training for Efficient Time Series Forecasting with Transformers
None
23-06-09
FeatureProgramming
ICML 2023
Feature Programming for Multivariate Time Series Prediction
FeatureProgramming
23-07-18
Look_Ahead
SIGIR 2023
Look Ahead: Improving the Accuracy of Time-Series Forecasting by Previewing Future Time Features
Look_Ahead
23-09-14
QFCV
Arxiv 2023
Uncertainty Intervals for Prediction Errors in Time Series Forecasting
QFCV
23-10-09
PeTS
Arxiv 2023
Performative Time-Series Forecasting
PeTS
23-10-23
EDAIN
Arxiv 2023
Extended Deep Adaptive Input Normalization for Preprocessing Time Series Data for Neural Networks
EDAIN
23-11-19
TimeSQL
Arxiv 2023
TimeSQL: Improving Multivariate Time Series Forecasting with Multi-Scale Patching and Smooth Quadratic Loss
None
24-01-16
LIFT
ICLR 2024
Rethinking Channel Dependence for Multivariate Time Series Forecasting: Learning from Leading Indicators
None
24-01-16
RobustTSF
ICLR 2024
RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies
RobustTSF
24-02-04
FreDF
Arxiv 2024
FreDF: Learning to Forecast in Frequency Domain
FreDF
24-02-20
Leddam
Arxiv 2024
Revitalizing Multivariate Time Series Forecasting: Learnable Decomposition with Inter-Series Dependencies and Intra-Series Variations Modeling
None
24-03-01
InfoTime
Arxiv 2024
Enhancing Multivariate Time Series Forecasting with Mutual Information-driven Cross-Variable and Temporal Modeling
None
24-03-13
wavelet-ML
Arxiv 2024
Leveraging Non-Decimated Wavelet Packet Features and Transformer Models for Time Series Forecasting
None
24-03-31
CCM
Arxiv 2024
From Similarity to Superiority: Channel Clustering for Time Series Forecasting
None
LLM (Large Language Model).
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
22-09-20
PromptCast
TKDE 2023
PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecasting
PISA
23-02-23
FPT π
NIPS 2023
One Fits All: Power General Time Series Analysis by Pretrained LM
One-Fits-All
23-05-17
LLMTime
NIPS 2023
Large Language Models Are Zero-Shot Time Series Forecasters
LLMTime
23-08-16
TEST
ICLR 2024
TEST: Text Prototype Aligned Embedding to Activate LLM's Ability for Time Series
None
23-08-16
LLM4TS
Arxiv 2023
LLM4TS: Two-Stage Fine-Tuning for Time-Series Forecasting with Pre-Trained LLMs
None
23-10-03
Time-LLM
ICLR 2024
Time-LLM: Time Series Forecasting by Reprogramming Large Language Models
None
23-10-08
TEMPO
ICLR 2024
TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting
None
23-10-12
Lag-Llama
Arxiv 2023
Lag-Llama: Towards Foundation Models for Time Series Forecasting
Lag-Llama
23-10-15
UniTime
Arxiv 2023
UniTime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecasting
None
23-11-03
ForecastPFN
NIPS 2023
ForecastPFN: Synthetically-Trained Zero-Shot Forecasting
ForecastPFN
23-11-24
FPT++ π
Arxiv 2023
One Fits All: Universal Time Series Analysis by Pretrained LM and Specially Designed Adaptors
GPT4TS_Adapter
24-01-18
ST-LLM
Arxiv 2024
Spatial-Temporal Large Language Model for Traffic Prediction
None
24-02-01
LLMICL
Arxiv 2024
LLMs learn governing principles of dynamical systems, revealing an in-context neural scaling law
LLMICL
24-02-04
AutoTimes
Arxiv 2024
AutoTimes: Autoregressive Time Series Forecasters via Large Language Models
AutoTimes
24-02-05
Position Paper
Arxiv 2024
Position Paper: What Can Large Language Models Tell Us about Time Series Analysis
None
24-02-16
TSFwithLLM
Arxiv 2024
Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities
None
24-02-25
LSTPrompt
Arxiv 2024
LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting
LSTPrompt
24-03-09
S2IP-LLM
Arxiv 2024
S2IP-LLM: Semantic Space Informed Prompt Learning with LLM for Time Series Forecasting
None
24-03-12
LLaTA
Arxiv 2024
Taming Pre-trained LLMs for Generalised Time Series Forecasting via Cross-modal Knowledge Distillation
LLaTA
24-03-12
Chronos π
Arxiv 2024
Chronos: Learning the Language of Time Series
Chronos
24-04-17
TSandLanguage
Arxiv 2024
Language Models Still Struggle to Zero-shot Reason about Time Series
TSandLanguage
Pretrain & Representation.
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
20-10-06
TST
KDD 2021
A Transformer-based Framework for Multivariate Time Series Representation Learning
mvts_transformer
21-09-29
CoST
ICLR 2022
CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecasting
CoST
22-05-16
LaST
NIPS 2022
LaST: Learning Latent Seasonal-Trend Representations for Time Series Forecasting
LaST
22-06-18
STEP
KDD 2022
Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting
STEP
23-02-02
SimMTM
NIPS 2023
SimMTM: A Simple Pre-Training Framework for Masked Time-Series Modeling
SimMTM
23-02-07
DBPM
ICLR 2024
Towards Enhancing Time Series Contrastive Learning: A Dynamic Bad Pair Mining Approach
None
23-03-01
TimeMAE
Arxiv 2023
TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders
TimeMAE
23-08-02
Floss
Arxiv 2023
Enhancing Representation Learning for Periodic Time Series with Floss: A Frequency Domain Regularization Approach
floss
23-12-01
STD_MAE
Arxiv 2023
Spatio-Temporal-Decoupled Masked Pre-training for Traffic Forecasting
STD_MAE
23-12-25
TimesURL
AAAI 2024
TimesURL: Self-supervised Contrastive Learning for Universal Time Series Representation Learning
None
24-01-08
TTMs
Arxiv 2024
TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series
None
24-01-16
SoftCLT
ICLR 2024
Soft Contrastive Learning for Time Series
None
24-01-16
PITS
ICLR 2024
Learning to Embed Time Series Patches Independently
PITS
24-01-16
T-Rep
ICLR 2024
T-Rep: Representation Learning for Time Series using Time-Embeddings
None
24-01-16
AutoTCL
ICLR 2024
Parametric Augmentation for Time Series Contrastive Learning
None
24-01-16
AutoCon
ICLR 2024
Self-Supervised Contrastive Forecasting
AutoCon
24-01-29
MLEM
Arxiv 2024
Self-Supervised Learning in Event Sequences: A Comparative Study and Hybrid Approach of Generative Modeling and Contrastive Learning
MLEM
24-02-04
Timer
Arxiv 2024
Timer: Transformers for Time Series Analysis at Scale
Timer
24-02-04
TimeSiam
Arxiv 2024
TimeSiam: A Pre-Training Framework for Siamese Time-Series Modeling
None
24-02-04
Moirai
Arxiv 2024
Unified Training of Universal Time Series Forecasting Transformers
Moirai
24-02-06
MOMENT
Arxiv 2024
MOMENT: A Family of Open Time-series Foundation Models
MOMENT
24-02-14
GTT
Arxiv 2024
Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction
GTT
24-02-26
TOTEM
Arxiv 2024
TOTEM: TOkenized Time Series EMbeddings for General Time Series Analysis
TOTEM
24-02-26
GPHT
Arxiv 2024
Generative Pretrained Hierarchical Transformer for Time Series Forecasting
None
24-02-29
UniTS
Arxiv 2024
UniTS: Building a Unified Time Series Model
UniTS
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
21-02-13
DAF
ICML 2022
Domain Adaptation for Time Series Forecasting via Attention Sharing
DAF
Date
Method
Type
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
22-02-23
FSNet
ICLR 2023
Learning Fast and Slow for Online Time Series Forecasting
FSNet
23-09-22
OneNet
NIPS 2023
OneNet: Enhancing Time Series Forecasting Models under Concept Drift by Online Ensembling
OneNet
23-09-25
MemDA
CIKM 2023
MemDA: Forecasting Urban Time Series with Memory-based Drift Adaptation
None
24-01-08
ADCSD
Arxiv 2024
Online Test-Time Adaptation of Spatial-Temporal Traffic Flow Forecasting
ADCSD
24-02-03
TSF-HD
Arxiv 2024
A Novel Hyperdimensional Computing Framework for Online Time Series Forecasting on the Edge
TSF-HD
24-02-20
SKI-CL
Arxiv 2024
Structural Knowledge Informed Continual Multivariate Time Series Forecasting
None
24-03-22
D3A
Arxiv 2024
Addressing Concept Shift in Online Time Series Forecasting: Detect-then-Adapt
None
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
22-10-25
WaveBound
NIPS 2022
WaveBound: Dynamic Error Bounds for Stable Time Series Forecasting
WaveBound
23-05-25
Ensembling
ICML 2023
Theoretical Guarantees of Learning Ensembling Strategies with Applications to Time Series Forecasting
None
Date
Method
Conference
Paper Title and Paper Interpretation (In Chinese)
Code
16-12-05
TRMF
NIPS 2016
Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction
TRMF
24-01-16
STanHop-Net
ICLR 2024
STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction
None
24-02-02
SNN
Arxiv 2024
Efficient and Effective Time-Series Forecasting with Spiking Neural Networks
None
24-03-12
BayesNF
Arxiv 2024
Scalable Spatiotemporal Prediction with Bayesian Neural Fields
BayesNF