Pith. sign in

REVIEW 3 major objections 5 minor 45 references

TLCCSP: A Scalable Framework for Enhancing Time Series Forecasting with Time-Lagged Cross-Correlations

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper argues that appending the top few time-lagged cross-correlated series to a forecaster's input lowers prediction error on weather, stock, and real estate data, and that a contrastive encoder can approximate the required correlatio

desk verdict Useful wrapper idea and a solid ablation, but the weather split leaks future information, so the headline MSE gains are not trustworthy as forecast improvements. read the letter →

arxiv 2508.07016 v1 pith:EORVSZ7T submitted 2025-08-09 cs.LG cs.IR

Jianfei Wu , Wenmian Yang , Bingning Liu , Weijia Jia This is my paper · ORCID
classification cs.LGcs.IR
keywords timeseriesforecastingtime-laggedcross-correlationdynamicwarpingcontrastivelearningauxiliarysequenceselectionSSDTWscalablesimilaritysearchaccuracy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a forecasting model can be improved simply by giving it, as extra input features, a small set of other time series that tend to move after a delay. The authors define a shift-aware similarity, SSDTW, as the minimum dynamic time warping distance between the target series and a candidate series shifted by a few days or months. Across weather, stock, and real estate data and seven forecasting backbones, adding the top $K_s$ series by this measure reduces mean squared error relative to forecasting from the target alone. To make the search scalable, the paper trains a contrastive learning encoder whose distances approximate SSDTW, and reports that this preserves most of the accuracy gain while cutting correlation-computation time by about 99%.

What carries the argument

The central mechanism is Sequence Shifted Dynamic Time Warping (SSDTW), defined as $\mathrm{SSDTW}(A,S)=\min_{\tau\in\mathcal{T}}\mathrm{DTW}(A,S_\tau)$, where $\mathcal{T}$ is a small domain-specific set of shift windows. It makes the similarity between two series depend on whether one moves after the other, not just on matched shape. The second mechanism is a contrastive learning encoder: a three-block convolutional network trained with the most SSDTW-correlated series as positives and the least as negatives, so that cosine distance in embedding space approximates SSDTW distance and replaces the expensive dynamic-programming search.

What would settle it

Run the same seven backbones on a held-out forecast period where all correlations are computed strictly from earlier data (weather: 2019 only; stock: through 2018 only; real estate: 2022 only). If Table 1's MSE reductions disappear or random selection matches them, the central claim fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that time-lagged cross-correlations between series are a reusable source of forecasting signal. It defines the Sequence Shifted Dynamic Time Warping distance as $\mathrm{SSDTW}(A,S)=\min_{\tau\in\mathcal{T}}\mathrm{DTW}(A,S_\tau)$, where $\mathcal{T}$ is a small set of shift windows chosen per dataset. Selecting the top $K_s$ candidate series by this distance and concatenating them to the target's history as auxiliary inputs lowers MSE relative to single-series forecasting: averages of 16.01% on weather, 9.95% on stock, and 21.29% on real estate across seven backbone models. The paper further claims a contrastive learning encoder whose embedding distances approxi

Load-bearing premise

The load-bearing premise is that the selected lag-correlated series carry information about the target's future that the target's own history does not already contain, and that the correlation periods used to select them do not leak future information into the forecast.

Editorial extensions

If this is right

  • Any existing forecaster can be upgraded by appending the top $K_s$ lag-correlated series to its input, with no change to the backbone architecture.
  • The contrastive encoder makes this selection cheap enough to update continuously on large candidate pools, enabling real-time retrieval of correlated series for financial and weather applications.
  • The framework is domain-agnostic: the same procedure works with day-level, multi-day, or month-level shift sets, as long as the shift set matches the data's timescale.
  • Choosing too many auxiliary series degrades accuracy, so the selection count $K_s$ is a meaningful hyperparameter rather than a free lunch.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper: the shift sets $\mathcal{T}$ are hand-picked per dataset; a natural extension is learning the shift set from data rather than choosing {1,3,5,10}, {5,10,20,30}, or {1,2,3}.
  • Beyond the paper: the reported gains are cell averages without variance or significance information, so a paired resampling analysis across random seeds would clarify whether the improvements are systematic or carried by a few backbones.
  • Beyond the paper: the paper notes CLE sometimes beats SSDTW selection, which suggests the embedding may encode more than the minimum-DTW score; testing whether embedding distances track lead-lag direction or correlation strength would clarify the mechanism.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes TLCCSP, a framework that augments a target time series with auxiliary series selected by a new SSDTW distance: the minimum DTW distance between the target and the candidate over a small set of time shifts. The top-Ks candidates are concatenated as extra input features to an arbitrary forecasting backbone. To avoid the O(N^2 T^2) cost of SSDTW, a contrastive learning encoder (CLE) is trained so that embedding distances approximate SSDTW distances, and the paper reports that using CLE cuts SSDTW computation by about 99%. Experiments on weather, stock, and real estate data with seven backbones report average MSE reductions of 16.01%, 9.95%, and 21.29% for SSDTW and 17.88%, 6.13%, and 8.62% for CLE relative to single-series forecasting. An ablation on the stock dataset compares SSDTW with random selection and plain DTW; sensitivity studies vary Ks, Ke, and the temperature lambda.

Significance. If the reported gains are real, the framework is a useful plug-in for existing forecasting models and the CLE speedup is practically significant. The ablation against random selection and plain DTW is a good control, and the case study supports the qualitative claim that SSDTW and CLE retrieve lagged correlated sequences. However, the main evidence is not yet convincing: on two of the three datasets the evaluation protocol uses a random split of overlapping sliding windows, which can leak the target into training and invalidate the reported reductions. The absence of any variance or significance information further weakens the quantitative claims. The work is therefore promising but requires a corrected evaluation before the central forecasting claim can be accepted. Strengths: the method is simple, model-agnostic, and the computational-cost reduction is explicitly measured; the paper does not oversell the SSDTW-vs-CLE comparison and even notes cases where CLE outperforms SSDTW.

major comments (3)
  1. [§4.1.1 and §4.1.3, Table 1] The weather and real estate datasets use sliding windows and then a random 6:2:2 train/validation/test split. Weather windows [1,50], [2,51], ... share 49 of 50 input days; real estate windows share 8 of 9 input months. A random split puts near-duplicate windows into different partitions, so the model can memorize the target rather than forecast it. The reported weather and real estate reductions (16.01% and 21.29%) are therefore not validated as out-of-sample improvements. A chronological split, as apparently used for the stock dataset, is required, and all three datasets should be evaluated under the same protocol.
  2. [Table 1, §4.1] No variance, number of runs, or significance statistics are reported for any entry. Many effects are small or negative (e.g., Stock LSTM SSDTW 1.68 vs Single 1.65; Real Estate TimesNet SSDTW 0.88 vs 0.87; Real Estate Transformer CLE 1.57 vs 1.55), so the average reductions in Table 1 cannot be distinguished from random seed variation. This is not a presentational issue: the central claim is quantitative, and the paper should provide repeated-run statistics or, at minimum, error bars for the headline numbers.
  3. [§3.1 Eq. (3), §3.2 Eq. (4)-(5)] SSDTW selects series by minimizing DTW over a set of shifts tau, but Eq. (3) feeds the raw auxiliary histories {S*_t} to the forecaster without the winning shift. Thus the framework does not actually use the estimated lag in prediction; it only uses the identity of the selected series. If the lag itself is informative, an experiment with shifted auxiliary inputs or explicit lag features is needed; if only the selection matters, the claim that TLCCSP 'captures time-lagged cross-correlations' should be softened. This distinction is central to the paper's stated mechanism.
minor comments (5)
  1. [Abstract, Table 1] The abstract says CLE 'further decreases' MSE on weather by 17.88%, but Table 1's Δ column compares CLE to Single, not to SSDTW. In stock and real estate, CLE is on average worse than SSDTW. Please rephrase to avoid implying a stacked improvement.
  2. [§4.4] The statement that SSDTW and CLE 'consistently rank first or second' is contradicted by Table 1: e.g., Stock LSTM SSDTW is worse than Single, and Real Estate TimesNet SSDTW is worse than Single. Please qualify the claim.
  3. [Eq. (6)] The positive and negative sample sets are described with the same index i and with conditions i ≤ Ke and i > N−Ke. This is ambiguous about the ordering of SSDTW distances; please define rank explicitly from 1 (nearest) to N (farthest).
  4. [Throughout] Typos and inconsistent terminology: 'TLCCSPP' in the conclusion, 'Conclution', 'appliations', 'sequencess', and 'candidate stock' in a weather context where the objects are cities. Please copyedit.
  5. [§3.3, §4.3] The CLE encoder is described only as a '3-block convolutional neural network' from ADATIME [40]; no layer sizes, pooling, or training hyperparameters for the encoder are given. Please provide enough detail for reproducibility.

Circularity Check

1 steps flagged · score 6.0 of 10

Weather (and likely real estate) evaluation randomly splits overlapping sliding windows, making reported test predictions in-sample; central MSE-reduction claims are not out-of-sample.

  1. fitted input called prediction [Section 4.1.1 (Weather Dataset), Section 4.1.3 (Real Estate Dataset), Table 1]
    "For the time series forecasting task, we use daily average temperature data from 2020 and follow the sliding window approach from FNSPID [15](e.g., [1,50], [2,51], ..., [T-49,T]). The first 49 days serve as inputs to predict the target at the 50th day. ... The 2020 dataset is randomly divided into training, validation, and test sets in a 6:2:2 ratio."

    With overlapping sliding windows, a random 6:2:2 split can place training window [1,50] (input days 1–49, target day 50) and test window [2,51] (input days 2–50, target day 51) in different partitions. The test input shares 48 of its 49 days with the training input and includes the training target day 50, so the forecaster can memorize near-duplicate sequences during training. The MSE reported in Table 1 is thus an in-sample fit, not an out-of-sample forecast; the claimed double-digit reductions on weather (16.01%) and real estate (21.29%, which uses the same random-partition protocol) are forced by the split rather than by the SSDTW/CLE selection.

full rationale

The framework itself is not self-referential: SSDTW distances are computed on historical periods disjoint from the forecast periods (2019 vs 2020 for weather; pre-2019 vs FNSPID test for stock; 2022 vs 2023 for real estate), the CLE is a supervised surrogate trained on SSDTW labels rather than an assumption of the result, and the forecasting comparison is a genuine input-augmentation experiment. The only self-citation (RETQA) supplies data, not a load-bearing theorem. However, the weather (and likely real estate) evaluation randomly splits overlapping sliding windows, making test windows near-duplicates of training windows and the reported 'predictions' in-sample. Since the abstract's headline MSE reductions rest on this protocol, the central empirical claim is circular in the fitted-input-called-prediction sense. Score is 6.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The framework introduces no new physical or mathematical entities. Its load-bearing components are the per-dataset shift sets and the retrieval hyperparameters Ks, Ke, and lambda, all tuned on the validation set or chosen by hand. The domain assumption that lagged correlations persist from the correlation period into the forecast period is the main premise that a reimplementation would need to check.

free parameters (4)
  • shift window set T = weather {1,3,5,10} days, stock {5,10,20,30} days, real estate {1,2,3} months
    Chosen per dataset by the authors in Section 4.3 without a data-driven criterion. The central claim depends on these selected lags.
  • Ks, number of selected correlated sequences = 3
    Set 'according to the optimal performance in the validation set' in Section 4.3 and varied in the sensitivity analysis of Section 4.6.
  • Ke, number of positive/negative samples for CLE = 5
    Set in Section 4.3 as 5 and varied in the sensitivity analysis of Section 4.6.
  • lambda, temperature in contrastive loss = 0.2
    Set in Section 4.3 as 0.2 and varied in the sensitivity analysis of Section 4.6.
assumptions (3)
  • domain assumption The correlated sequences selected by SSDTW on historical data remain relevant for the forecasting target period.
    The framework computes correlations on one period (2019 weather, pre-2018 stocks, 2022 real estate) and applies the selected relationships to forecast a later period. If the lag structure changes over time, the auxiliary information could degrade.
  • domain assumption Time-lagged cross-correlations exist and are exploitable in the chosen datasets and horizons.
    The method is motivated by cited work on weather and finance, and the experimental results support it, but the paper postulates that such correlations are present and useful rather than measuring them independently before the experiments.
  • ad hoc to paper The DTW distance over a small set of shifts is a good proxy for predictive relevance of the auxiliary series.
    SSDTW rank is used to select which series help forecasting, but the paper does not establish that similarity under DTW after shifting is the optimal criterion for predictive gain. The ablation shows it beats random and plain DTW, but not that it is the right criterion.

how reviews work

0 comments
Cite this review

Pith. "Pith review of TLCCSP: A Scalable Framework for Enhancing Time Series Forecasting with Time-Lagged Cross-Correlations." pith.science (2026). https://pith.science/paper/EORVSZ7T

@misc{pith2026250807016,
  author       = {Pith},
  title        = {Pith review of: TLCCSP: A Scalable Framework for Enhancing Time Series Forecasting with Time-Lagged Cross-Correlations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EORVSZ7T}},
  note         = {Machine review of arXiv:2508.07016}
}
read the original abstract

Time series forecasting is critical across various domains, such as weather, finance and real estate forecasting, as accurate forecasts support informed decision-making and risk mitigation. While recent deep learning models have improved predictive capabilities, they often overlook time-lagged cross-correlations between related sequences, which are crucial for capturing complex temporal relationships. To address this, we propose the Time-Lagged Cross-Correlations-based Sequence Prediction framework (TLCCSP), which enhances forecasting accuracy by effectively integrating time-lagged cross-correlated sequences. TLCCSP employs the Sequence Shifted Dynamic Time Warping (SSDTW) algorithm to capture lagged correlations and a contrastive learning-based encoder to efficiently approximate SSDTW distances. Experimental results on weather, finance and real estate time series datasets demonstrate the effectiveness of our framework. On the weather dataset, SSDTW reduces mean squared error (MSE) by 16.01% compared with single-sequence methods, while the contrastive learning encoder (CLE) further decreases MSE by 17.88%. On the stock dataset, SSDTW achieves a 9.95% MSE reduction, and CLE reduces it by 6.13%. For the real estate dataset, SSDTW and CLE reduce MSE by 21.29% and 8.62%, respectively. Additionally, the contrastive learning approach decreases SSDTW computational time by approximately 99%, ensuring scalability and real-time applicability across multiple time series forecasting tasks.

Figures

Figures reproduced from arXiv: 2508.07016 by the authors.

Figure 1
Figure 1. Overall Framework of TLCCSP. auxiliary information in developing robust and reliable prediction models, further emphasizing the importance of comprehensive data integration in time series forecasting. In addition to the factors mentioned above, studies in weather forecasting and financial analysis [9–11] have identified time-lagged cross-correlations (TLCC) between individual sequences. For in￾stance, temperature ch… view at source ↗
Figure 2
Figure 2. MSE and MAE of Different models on weather dataset with different settings of [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. MSE and MAE of Different models on weather dataset with different settings of [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: MSE and MAE of Different models on weather dataset with different settings of [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Case Study: Correlated Sequences Selected by SSDTW and CLE [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 40 canonical work pages

  1. [1]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention Is All You Need. Advances in Neural Information Processing Systems 30 (2017), 5998– 6008

    work page 2017

  2. [2]

    Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang. 2021. Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35. 11106–11115

    work page 2021

  3. [3]

    Haixu Wu, Tengge Hu, Yong Liu, Hang Zhou, Jianmin Wang, and Mingsheng Long. 2023. TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. In The Eleventh International Conference on Learning Representations . https://openreview.net/forum?id=ju_Uqw384Oq

    work page 2023

  4. [4]

    Yong Liu, Tengge Hu, Haoran Zhang, Haixu Wu, Shiyu Wang, Lintao Ma, and Mingsheng Long. 2023. Itransformer: Inverted Transformers Are Effective for Time Series Forecasting. arXiv Preprint arXiv:2310.06625 (2023)

    arXiv 2023

  5. [5]

    Kevin Höhlein, Benedikt Schulz, Rüdiger Westermann, and Sebastian Lerch. 2024. Postprocessing of Ensemble Weather Forecasts Using Permutation-Invariant Neural Networks. Artificial Intelligence for Earth Systems 3, 1 (2024), e230070. doi:10.1175/AIES-D-23-0070.1

    doi:10.1175/aies-d-23-0070.1 2024

  6. [6]

    L. J. Slater, L. Arnal, M. Boucher, A. Y.-Y. Chang, S. Moulds, C. Murphy, G. Nearing, G. Shalev, C. Shen, L. Speight, G. Villarini, R. L. Wilby, A. Wood, and M. Zappa. 2023. Hybrid Forecasting: Blending Climate Predictions with AI Models. Hydrology and Earth System Sciences 27, 9 (2023), 1865–1889. doi:10.5194/hess-27-1865-2023

    doi:10.5194/hess-27-1865-2023 2023

  7. [7]

    Md Masum Billah, Azmery Sultana, Farzana Bhuiyan, and Mohammed Golam Kaosar. 2024. Stock Price Prediction: Comparison of Different Moving Average Techniques Using Deep Learning Model. Neural Computing and Applications 36, 11 (2024), 5861–5871

    work page 2024

  8. [8]

    Zen-Yu Quan. 2013. Stock Prediction by Searching Similar Candlestick Charts. In 2013 IEEE 29th International Conference on Data Engineering Workshops (ICDEW) . IEEE, 322–325

    work page 2013

Show all 45 references
  1. [9]

    Haoliang Wang, Yubao Liu, Tianliang Zhao, Yuewei Liu, Mei Xu, Si Shen, Yin Jiang, Honglong Yang, and Shuanglei Feng. 2018. Continuous Assimilation of Lightning Data Using Time-Lagged Ensembles for a Convection-Allowing Nu- merical Weather Prediction Model. Journal of Geophysic...

    2018

  2. [10]

    Borysov and Alexander V

    Stanislav S. Borysov and Alexander V. Balatsky. 2014. Cross-Correlation Asym- metries and Causal Relationships Between Stock and Market Risk. PLoS One 9, 8 (2014), e105874

    2014

  3. [11]

    Jose Arreola Hernandez, Mazin A. M. Al Janabi, Shawkat Hammoudeh, and Duc Khuong Nguyen. 2015. Time Lag Dependence, Cross-Correlation and Risk Analysis of US Energy and Non-Energy Stock Portfolios. Journal of Risk and Financial Management 16 (2015), 467–483

    2015

  4. [12]

    Zhao Lv, Bei-Bei Zhang, Xiao-Pei Wu, Chao Zhang, and Bang-Yan Zhou. 2017. A Permutation Algorithm Based on Dynamic Time Warping in Speech Frequency- Domain Blind Source Separation. Journal of Information Processing 92 (2017), 132–141

    2017

  5. [13]

    Radu-Sebastian Marinescu, Alexandru George Rusu, Corneliu Burileanu, and Dumitru Bica. 2019. Simultaneous Speech Detection Based on MFCC-DTW with Two-Stage Normalization. Journal of Engineering Science and Technology 8 (2019), 29–29

    2019

  6. [14]

    NOAA. [n. d.]. NOAA Weather Dataset. ([n. d.]). https://www.ncei.noaa.gov/ data/local-climatological-data/

  7. [15]

    Zihan Dong, Xinyu Fan, and Zhiyuan Peng. 2024. Fnspid: A Comprehensive Financial News Dataset in Time Series. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining . 4918–4927

    2024

  8. [16]

    Zhensheng Wang, Wenmian Yang, Kun Zhou, Yiquan Zhang, and Weijia Jia. 2024. RETQA: A Large-Scale Open-Domain Tabular Question Answering Dataset for Real Estate Sector. In The 39th Annual AAAI Conference on Artificial Intelligence . https://arxiv.org/abs/2412.10104

    2024 arXiv

  9. [17]

    Anderson and M

    Oliver D. Anderson and M. G. Kendall. 1976. Time-Series. The Statistician 25 (1976), 308

    1976

  10. [18]

    Ponce Flores, J

    M. Ponce Flores, J. Frausto Solis, G. Castilla Valdez, J. J. Gonzalez Barbosa, J. Perez Ortega, and J. D. Teran Villanueva. 2019. Hurst Exponent with ARIMA and Simple Exponential Smoothing for Measuring Persistency of M3-Competition Series. Revista IEEE América Latina 17, 5 (2...

    2019

  11. [19]

    Jr. E. S. Gardner. 1985. Exponential Smoothing: The State of the Art. Journal of Forecasting 4 (1985), 1–28

    1985

  12. [20]

    Cansu, E

    T. Cansu, E. Bas, E. Egrioglu, and T. Akkan. 2024. Intuitionistic Fuzzy Time Series Forecasting Method Based on Dendrite Neuron Model and Exponential Smoothing. Granular Computing 9, 2 (2024). doi:10.1007/s41066-024-00474-6

    2024 doi

  13. [21]

    Haixu Wu, Tengge Hu, Yong Liu, Hang Zhou, Jianmin Wang, and Mingsheng Long. 2023. TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. In International Conference on Learning Representations

    2023

  14. [22]

    Bryan Lim and Stefan Zohren. 2021. Time-Series Forecasting with Deep Learning: A Survey. Philosophical Transactions of the Royal Society A 379, 2194 (2021), 20200209

    2021

  15. [23]

    Haugsdal, E

    E. Haugsdal, E. Aune, and M. Ruocco. 2023. Persistence Initialization: A Novel Adaptation of the Transformer Architecture for Time Series Forecasting. Applied Intelligence 53, 22 (2023), 26781–26796. doi:10.1007/s10489-023-04927-4

    2023 doi

  16. [24]

    Stefano Frizzo Stefenon, Laio Oriel Seman, Evandro Cardozo da Silva, Erlon Cris- tian Finardi, Leandro dos Santos Coelho, and Viviana Cocco Mariani. 2024. Hy- pertuned Wavelet Convolutional Neural Network with Long Short-Term Memory for Time Series Forecasting in Hydroelectric...

    2024

  17. [25]

    Carmina Fjellström. 2022. Long Short-Term Memory Neural Network for Finan- cial Time Series. In 2022 IEEE International Conference on Big Data (Big Data) . 3496–3504. doi:10.1109/BigData55660.2022.10020784

    2022

  18. [26]

    D. Wang, C. Chen, and L. Qi et al. 2023. Spatiotemporal Self-Attention-Based LSTNet for Multivariate Time Series Prediction.International Journal of Intelligent Systems (2023), 1–16. doi:10.1155/2023/9523230

    2023 doi

  19. [27]

    Xiaoli Ren, Xiaoyong Li, Kaijun Ren, Junqiang Song, Zichen Xu, Kefeng Deng, and Xiang Wang. 2021. Deep Learning-Based Weather Prediction: A Survey. Big Data Research 23 (2021), 100178

    2021

  20. [28]

    Ross Koval, Nicholas Andrews, and Xifeng Yan. 2023. Forecasting Earnings Surprises from Conference Call Transcripts. Findings of the Association for Computational Linguistics: ACL 2023 1 (2023), 8197–8209

    2023

  21. [29]

    Thuraiya Mohd, Nur Syafiqah Jamil, Noraini Johari, Lizawati Abdullah, and Suraya Masrom. 2020. An overview of real estate modelling techniques for house price prediction. In Charting a Sustainable Future of ASEAN in Business and Social Sciences: Proceedings of the 3rd Internat...

    2020

  22. [30]

    Jianhua Dong, Wenzhi Zeng, Lifeng Wu, Jiesheng Huang, Thomas Gaiser, and Amit Kumar Srivastava. 2023. Enhancing Short-Term Forecasting of Daily Pre- cipitation Using Numerical Weather Prediction Bias Correcting with XGBoost in Different Regions of China. Engineering Applicatio...

    2023

  23. [31]

    Jiaqi Zheng, Qing Ling, Jia Li, and Yerong Feng. 2024. Improving the Short- Range Precipitation Forecast of Numerical Weather Prediction through a Deep Learning-Based Mask Approach. Advances in Atmospheric Sciences 41, 8 (2024), 1601–1613

    2024

  24. [32]

    Kaifeng Bi, Lingxi Xie, Hengheng Zhang, Xin Chen, Xiaotao Gu, and Qi Tian

  25. [33]

    Yi-Ling Hsu, Yu-Che Tsai, and Cheng-Te Li. 2021. FinGAT: Financial Graph At- tention Networks for Recommending Top-𝑘 Profitable Stocks. IEEE Transactions on Knowledge and Data Engineering 35, 1 (2021), 469–481

    2021

  26. [34]

    Lu Bai, Lixin Cui, Zhihong Zhang, Lixiang Xu, Yue Wang, and Edwin R. Hancock

  27. [35]

    Guowei Song, Tianlong Zhao, Suwei Wang, Hua Wang, and Xuemei Li. 2023. Stock Ranking Prediction Using a Graph Aggregation Network Based on Stock Price and Stock Relationship Information.Information Sciences 643 (2023), 119236

    2023

  28. [36]

    Yaping Zhao, Jichang Zhao, and Edmund Y Lam. 2024. House price prediction: A multi-source data fusion perspective. Big Data Mining and Analytics 7, 3 (2024), 603–620

    2024

  29. [37]

    Sheng-Min Chiu, Yi-Chung Chen, and Chiang Lee. 2022. Estate price prediction system based on temporal and spatial features and lightweight deep learning model. Applied Intelligence 52, 1 (2022), 808–834

    2022

  30. [38]

    Duo Liu, Xingni Li, Po-Chun Huang, Yi Gu, Yingjian Ling, Kan Zhong, Renping Liu, Xianzhang Chen, and Liang. 2020. Downsizing Without Downgrading: Approximated Dynamic Time Warping on Nonvolatile Memories. Journal of CIKM, Nov 10–14, 2025, COEX, SEOUL, KOREA Wu et al. Non-Cryst...

    2020

  31. [39]

    Afshar Alam

    Munshi Yadav and M. Afshar Alam. 2018. Dynamic Time Warping (DTW) Algorithm in Speech: A Review. International Journal of Research in Electronics and Computer Engineering 6, 1 (2018), 524–528

    2018

  32. [40]

    Mohamed Ragab, Emadeldeen Eldele, Wee Ling Tan, Chuan-Sheng Foo, Zhenghua Chen, Min Wu, Chee-Keong Kwoh, and Xiaoli Li. 2023. ADATIME: A Bench- marking Suite for Domain Adaptation on Time Series Data. ACM Transactions on Knowledge Discovery from Data (2023). doi:10.1145/3587937

    2023 doi

  33. [41]

    Jiuxiang Gu, Zhenhua Wang, Jason Kuen, Lianyang Ma, Amir Shahroudy, Bing Shuai, Ting Liu, Xingxing Wang, and Gang Wang. 2018. Recent Advances in Convolutional Neural Networks. Pattern Recognition 77 (2018), 354–377

    2018

  34. [42]

    Medsker, Lakhmi Jain, et al

    Larry R. Medsker, Lakhmi Jain, et al. 2001. Recurrent Neural Networks. Design and Applications 5, 64–67 (2001), 2

    2001

  35. [43]

    Alex Sherstinsky. 2020. Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network. Physica D: Nonlinear Phenomena 404 (2020), 132306

    2020

  36. [2020]

    IEEE Transactions on Neural Networks and Learning Systems 34, 4 (2020), 1808–1822

    Entropic Dynamic Time Warping Kernels for Co-evolving Financial Time Series Analysis. IEEE Transactions on Neural Networks and Learning Systems 34, 4 (2020), 1808–1822

    2020

  37. [2023]

    Nature 619, 7970 (2023), 533–538

    Accurate Medium-Range Global Weather Forecasting with 3D Neural Networks. Nature 619, 7970 (2023), 533–538

    2023

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.