Financial Time Series Forecasting using CNN and Transformer
Proposes combining CNNs and Transformers for financial time series forecasting, outperforming traditional and deep learning methods.
What it examines
This paper addresses the challenge of financial time series forecasting by combining Convolutional Neural Networks (CNN) and Transformers to model both short-term and long-term dependencies. The proposed method aims to predict intraday stock price changes of S&P 500 constituents, outperforming traditional statistical and deep learning methods.
What it concludes
The study demonstrates the combined power of CNN and Transformer models for financial time series forecasting, showing potential for downstream trading decisions. Future research could explore further applications and improvements of this method in financial markets.
Evidence objects
The study demonstrates the combined power of CNN and Transformer models for financial time series forecasting, showing potential for downstream trading decisions. Future research could explore further applications and improvements of this method in financial markets.
key_findings bullet 1 · key_findings · validation V0
Raw abstract and provenance
Abstract: Time series forecasting is important across various domains for decision-making. In particular, financial time series such as stock prices can be hard to predict as it is difficult to model short-term and long-term temporal dependencies between data points. Convolutional Neural Networks (CNN) are good at capturing local patterns for modeling short-term dependencies. However, CNNs cannot learn long… ▽ More Time series forecasting is important across various domains for decision-making. In particular, financial time series such as stock prices can be hard to predict as it is difficult to model short-term and long-term temporal dependencies between data points. Convolutional Neural Networks (CNN) are good at capturing local patterns for modeling short-term dependencies. However, CNNs cannot learn long-term dependencies due to the limited receptive field. Transformers on the other hand are capable of learning global context and long-term dependencies. In this paper, we propose to harness the power of CNNs and Transformers to model both short-term and long-term dependencies within a time series, and forecast if the price would go up, down or remain the same (flat) in the future. In our experiments, we demonstrated the success of the proposed method in comparison to commonly adopted statistical and deep learning methods on forecasting intraday stock price change of S&P 500 constituents. △ Less
Source row: 825 · abstract type: unknown