Summary of Double-path Adaptive-correlation Spatial-temporal Inverted Transformer For Stock Time Series Forecasting, by Wenbo Yan and Ying Tan
Double-Path Adaptive-correlation Spatial-Temporal Inverted Transformer for Stock Time Series Forecasting
by Wenbo Yan, Ying Tan
First submitted to arxiv on: 24 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a new spatial-temporal graph neural network (STGNN) called DPA-STIFormer for improving stock prediction tasks. Unlike existing STGNNs that learn spatial relationships from time series, this model explicitly models feature changes as tokens and extracts dynamic spatial information from stock data. The Double-Path Adaptive-correlation Spatial-Temporal Inverted Transformer (DPA-STIFormer) uses a novel fusion mechanism to combine temporal and feature representations, allowing it to uncover latent temporal-correlation patterns. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Stock prediction is tricky because traditional methods don’t capture the complex relationships between stocks. Researchers tried using special kinds of neural networks called STGNNs, but they still didn’t do well. This paper suggests a new kind of STGNN that looks at how features change over time, rather than just looking at the time series itself. It’s like taking a closer look at how the numbers move instead of just looking at the numbers themselves. The new model is called DPA-STIFormer and it uses a special way to combine different kinds of information to make predictions. |
Keywords
» Artificial intelligence » Graph neural network » Time series » Transformer
