TCGPN: Temporal-Correlation Graph Pre-trained Network for Stock Forecasting

TL;DR

This paper introduces TCGPN, a novel pre-trained graph neural network that effectively captures temporal correlations in stock data, especially for non-periodic datasets, achieving state-of-the-art forecasting results with reduced memory usage.

Contribution

The paper proposes TCGPN, a pre-trained network that handles large-scale, non-periodic stock data by using a temporal-correlation encoder and a node-independent structure, improving robustness and efficiency.

Findings

Achieves state-of-the-art results on CSI300 and CSI500 datasets.

Effectively captures robust temporal correlation patterns.

Reduces memory consumption during training.

Abstract

Recently, the incorporation of both temporal features and the correlation across time series has become an effective approach in time series prediction. Spatio-Temporal Graph Neural Networks (STGNNs) demonstrate good performance on many Temporal-correlation Forecasting Problem. However, when applied to tasks lacking periodicity, such as stock data prediction, the effectiveness and robustness of STGNNs are found to be unsatisfactory. And STGNNs are limited by memory savings so that cannot handle problems with a large number of nodes. In this paper, we propose a novel approach called the Temporal-Correlation Graph Pre-trained Network (TCGPN) to address these limitations. TCGPN utilize Temporal-correlation fusion encoder to get a mixed representation and pre-training method with carefully designed temporal and correlation pre-training tasks. Entire structure is independent of the number and order of nodes, so better results can be obtained through various data enhancements. And memory consumption during training can be significantly reduced through multiple sampling. Experiments are conducted on real stock market data sets CSI300 and CSI500 that exhibit minimal periodicity. We fine-tune a simple MLP in downstream tasks and achieve state-of-the-art results, validating the capability to capture more robust temporal correlation patterns.