Abstract:Market financial forecasting is a trending area in deep learning. Deep learning models are capable of tackling the classic challenges in stock market data, such as its extremely complicated dynamics as well as long-term temporal correlation. To capture the temporal relationship among these time series, recurrent neural networks are employed. However, it is difficult for recurrent models to learn to keep track of long-term information. Convolutional Neural Networks have been utilized to better capture the dynamics and extract features for both short- and long-term forecasting. However, semantic segmentation and its well-designed fully convolutional networks have never been studied for time-series dense classification. We present a novel approach to predict long-term daily stock price change trends with fully 2D-convolutional encoder-decoders. We generate input frames with daily prices for a time-frame of T days. The aim is to predict future trends by pixel-wise classification of the current price frame. We propose a hierarchical CNN structure to encode multiple price frames to multiscale latent representation in parallel using Atrous Spatial Pyramid Pooling blocks and take that temporal coarse feature stacks into account in the decoding stages. Our hierarchical structure of CNNs makes it capable of capturing both long and short-term temporal relationships effectively. The effect of increasing the input time horizon via incrementing parallel encoders has been studied with interesting and substantial changes in the output segmentation masks. We achieve overall accuracy and AUC of %78.18 and 0.88 for joint trend prediction over the next 20 days, surpassing other semantic segmentation approaches. We compared our proposed model with several deep models specifically designed for technical analysis and found that for different output horizons, our proposed models outperformed other models.

What problem does this paper attempt to address?

The paper aims to address the problem of stock price trend prediction, particularly by leveraging deep learning techniques to improve the accuracy of both long-term and short-term trend predictions. The authors propose a novel approach that applies semantic segmentation techniques to time series data to predict changes in stock price trends. Specifically, the main contributions of the paper include: 1. **Proposing a 2D price matrix**: A 2D matrix is formed by the daily opening price, highest price, lowest price, and closing price, simulating the input style of an image for the semantic segmentation task. The final segmentation mask corresponds to the price trends over the next T days. 2. **Designing a fully convolutional encoder-decoder network**: This network can effectively extract and integrate short-term and long-term patterns. By using Atrous Spatial Pyramid Pooling (ASPP) blocks, the network can extract patterns at different receptive fields and obtain deep coarse-grained feature stacks from multiple consecutive price frames through multi-level downsampling. After fusing the extracted features, these deep coarse-grained feature maps are upsampled to generate a complete segmentation map, which is used to predict the price change direction for the next T days. 3. **Experimental validation**: Through the study of different input time ranges, it was found that as the input time range increases, the overall results improve, and the most accurate prediction date gradually moves to the middle of the output time range. Additionally, compared to previous recursive and feedforward models, the proposed network shows significant advantages in trend prediction performance evaluation and semantic segmentation evaluation. In summary, this paper proposes a novel semantic segmentation-based method to predict stock price trends and demonstrates its effectiveness through experiments. This method not only improves prediction accuracy but also showcases the advantages of using convolutional neural networks in handling time series data.

Stock Trend Prediction: A Semantic Segmentation Approach

Stock Price Prediction using Multi-Faceted Information based on Deep Recurrent Neural Networks

Research on a Trend Forecast Model of Time Series Data Based on Deep Learning

A Stock Price Prediction Approach Based on Time Series Decomposition and Multi-Scale CNN using OHLCT Images

Stock Market Prediction Based on Historic Prices and News Titles

A Time Series Analysis-Based Stock Price Prediction Using Machine Learning and Deep Learning Models

Deep learning framework for stock price prediction using long short-term memory

Sector influence aware stock trend prediction using 3D convolutional neural network

Stock Market Trend Prediction Using High-Order Information of Time Series

Stock Price Prediction based on Recurrent Neural Network with Long Short-Term Memory Units

Support for Stock Trend Prediction Using Transformers and Sentiment Analysis

An Evaluation of Deep Learning Models for Stock Market Trend Prediction

A Stock Prediction Model Based on DCNN

Stock market forecasting using deep learning with long short-term memory and gated recurrent unit

Deep learning for Stock Market Prediction

Stock Price Prediction Using Convolutional Neural Networks on a Multivariate Timeseries

A Hybrid Data Analytics Framework with Sentiment Convergence and Multi-Feature Fusion for Stock Trend Prediction

Predicting Stock Market Trends with a Transfer Learning Model Based on CNN-LSTM Architecture

Hierarchical Adaptive Temporal-Relational Modeling for Stock Trend Prediction

Stock Price Prediction and Traditional Models: An Approach to Achieve Short-, Medium- and Long-Term Goals

HATR-I: Hierarchical Adaptive Temporal Relational Interaction for Stock Trend Prediction