STP-TrellisNets+: Spatial-Temporal Parallel TrellisNets for Multi-Step Metro Station Passenger Flow Prediction

Ou, Junjie; Sun, Jiahui; Zhu, Yichen; Jin, Haiming; Liu, Yijuan; Zhang, Fan; Huang, Jianqiang; Wang, Xinbing

doi:10.1109/tkde.2022.3187690

Cited by 10 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A number of scholars have recently exploited dynamic spatiotemporal correlations of urban traffic by proposing models based on machine learning. For example, Ou et al (2022) proposed a novel deep-learning framework, STP-TrellisNets, to predict passenger flows at metro stations. Liu et al (2021) forecast future citywide crowd flows to facilitate urban management by modeling spatiotemporal patterns of recent crowd flows.…”

Section: Literature Review and Model Designmentioning

confidence: 99%

“…Spatial and temporal correlations also shift over time; using static cross-sectional data solely for analysis generates biased results. Lately, several papers have addressed dynamic spatiotemporal correlations via machine learning (Ou et al 2022;Liu et al 2021;Pan et al 2022). However, these models cannot be applied directly because they describe certain spatiotemporal characteristics of citywide traffic flow but not financial issues.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic spatiotemporal correlation coefficient based on adaptive weight

Tan

Zhang

et al. 2023

Financ Innov

View full text Add to dashboard Cite

Risk management is an important aspect of financial research because correlations among financial data are essential in evaluating portfolio risk. Among various correlations, spatiotemporal correlations involve economic entity attributes and are interrelated in space and time. Such correlations have therefore drawn increasing attention in financial risk management. However, classical correlation measurements are typically based on either time series correlations or spatial dependence; they cannot be directly applied to financial data with spatiotemporal correlations. The spatiotemporal correlation coefficient model with adaptive weight proposed in this paper can (1) address the absolute quantity, dynamic quantity, and dynamic development of financial data and (2) be used for risk grading, financial risk evaluation, and portfolio management. To verify the validity and superiority of this model, cluster analysis results and portfolio performance are compared with a classical model with time series correlation or spatial correlation, respectively. Empirical findings show that the proposed coefficient is highly effective and convenient compared to others. Overall, our method provides a highly efficient financial risk management method with valuable implications for investors and financial institutions.

show abstract

Section: Literature Review and Model Designmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Dynamic spatiotemporal correlation coefficient based on adaptive weight

Tan

Zhang

et al. 2023

Financ Innov

View full text Add to dashboard Cite

show abstract

“…Multi-step rolling forecasting strategy: Rolling forecasting strategy [4], [58], [59] uses (T − t) steps to forecast the whole prediction window. kth step forecasts the elements at time stamp (t + k) with input sequence window…”

Section: Appendix a Preliminarymentioning

confidence: 99%

“…With the growing demand of long-term time series forecasting accuracy in various domains [8], [58], [60], [61], [62], [63], [64], [65], [66], traditional forecasting models, e.g. ARIMA [20], [21], SES [67], are no longer able to deal with more and more complicated forecasting situations.…”

Section: Appendix B Related Workmentioning

confidence: 99%

Respecting Time Series Properties Makes Deep Time Series Forecasting Perfect

Shen¹,

Wei²,

Wang³

2022

Preprint

View full text Add to dashboard Cite

How to handle time features shall be the core question of any time series forecasting model. Ironically, it is often ignored or misunderstood by deep-learning based models, even those baselines which are state-of-the-art. This behavior makes their inefficient, untenable and unstable. In this paper, we rigorously analyze three prevalent but deficient/unfounded deep time series forecasting mechanisms or methods from the view of time series properties, including normalization methods, multivariate forecasting and input sequence length. Corresponding corollaries and solutions are given on both empirical and theoretical basis. We thereby propose a novel time series forecasting network, i.e. RTNet, on the basis of aforementioned analysis. It is general enough to be combined with both supervised and self-supervised forecasting format. Thanks to the core idea of respecting time series properties, no matter in which forecasting format, RTNet shows obviously superior forecasting performances compared with dozens of other SOTA time series forecasting baselines in three real-world benchmark datasets. By and large, it even occupies less time complexity and memory usage while acquiring better forecasting accuracy. The source code is available at https://github.com/OrigamiSL/RTNet.

show abstract

“…An end-to-end global spatial-temporal graph attention network (GST-GAT) proposed by [31] uses "global interaction + node query" to model the dynamic spatial-temporal relations of traffic data. Furthermore, [32] embedded attention mechanism into GCN to obtain dynamic spatial features by assigning a probability to a road segment so as to contribute to the targeted road segment for traffic prediction, while [33] designed a novel structure based on GCN, named GC-TrellisNetsED, to capture the spatial correlation among metro stations and also the dynamics of such correlations, working with a temporal module for multi-step metro station passenger flow prediction.…”

Section: Introductionmentioning

confidence: 99%