Summary of Tcgpn: Temporal-correlation Graph Pre-trained Network For Stock Forecasting, by Wenbo Yan et al.
TCGPN: Temporal-Correlation Graph Pre-trained Network for Stock Forecasting
by Wenbo Yan, Ying Tan
First submitted to arxiv on: 26 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Statistical Finance (q-fin.ST); Machine Learning (stat.ML)
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 This paper proposes a novel approach called Temporal-Correlation Graph Pre-trained Network (TCGPN) for time series prediction, which addresses limitations of Spatio-Temporal Graph Neural Networks (STGNNs). TCGPN utilizes Temporal-correlation fusion encoder to capture robust temporal correlation patterns and pre-training method with carefully designed tasks. The model is independent of node order and number, making it suitable for various data enhancements and reducing memory consumption during training. Experimental results on real stock market datasets CSI300 and CSI500 demonstrate state-of-the-art performance, validating the model’s capability to capture temporal correlations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to predict things that happen over time. Right now, some models are good at predicting patterns in data that change over time, but they struggle when there’s no clear pattern. This new approach, called TCGPN, is better at finding hidden patterns and can handle lots of data without getting slow or running out of memory. The researchers tested it on real stock market data and got the best results ever! |
Keywords
* Artificial intelligence * Encoder * Time series
