Stock Dataset Research Articles

Change point detection via feedforward neural networks with theoretical guarantees

This article mainly studies change point detection for mean shift change point model. An estimation method is proposed to estimate the change point via feedforward neural networks. The complete f-moment consistency of the proposed estimator is obtained. Numerical simulation results show that the performance of the proposed estimator is better than that of cumulative sum type estimator which is widely used in the change point detection, especially when the mean shift signal size is small. Finally, we demonstrate the proposed method by empirically analyzing a stock data set.

Federated Meta Embedding Concept Stock Recommendation

Mining relevant stocks given a trending topic/concept in capital markets is an application with significant economic and societal impacts. Previous concept stock recommendation system mines concept stocks only from public social media like financial news. On stock forums, investors discuss emerging concepts and stocks by using forum comments, which are unneglectable resources to capture trending concept stocks accurately and timely. However, the comment data from a single forum is insufficient to build a high-quality recommendation system. The forums are data silos protected by privacy regulations, and their comments are still underutilized. In this paper, we propose a federated concept stock recommendation baseline and an optimized method that both leverage the private forum comments and public social media without compromising privacy regulations. Our baseline, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , Federated Meta Embedding (FedME), is built upon the federated learning framework and learns a concept-stock embedding jointly from private and public data. Our optimized method, Federated Graph Meta Embedding (FedGME), improves FedME by using a graph to combine two sources of embeddings and additional human experts' concept-stock knowledge. Empirically, the experiments on two concept stock datasets show that FedME and FedGME substantially improve the performance of recommendation. Our methods provide practical guidance on privacy-preserving FinTech applications.

Deep Learning Approaches for Predicting Intraday Price Movements: An Evaluation of RNN Variants on High-Frequency Stock Data

This study discusses the comparison of four recurrent neural networks (RNN) models: Simple RNN, Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Bidirectional RNN (BiRNN), in forecasting minute-level stock price time series data. The performance of these four models is evaluated using the Mean Absolute Percentage Error (MAPE) on a stock dataset from Bank Central Asia (BBCA.JK). The experimental results reveal that the GRU model exhibits the best performance with an average MAPE of 0.0255%, followed by the LSTM model with an average MAPE of 0.0377%. The BiRNN model also demonstrates good performance with an average MAPE of 0.0668%, while the Simple RNN has the highest average MAPE at 0.5118%. This suggests that more complex recurrent architectures like GRU and LSTM have better capabilities in capturing patterns in high-frequency time series data. This study can be expanded by exploring other models such as CNN, conducting tests on diverse datasets, and experimenting with a wider range of hyperparameter variations. Additional variables such as economic indicators, global market data, and social data can also offer a more comprehensive understanding of factors influencing stock prices.

Dual-Attention Based Multi-Path Approach for Intensifying Stock Market Forecasting

In light of the existing challenges in capturing short-term fluctuations and achieving accurate predictions in stock market time series data, this research presents the “Dual-Attention MDWT-CVT-LSTM,” a revolutionary financial time series forecasting model. This model makes use of a dual-attention mechanism in conjunction with a Variant Transformation-Gated Long Short-Term Memory (LSTM) network. The approach relies on the Multilevel Discrete Wavelet Transform (MDWT) to separate stock market index sequences into high-frequency and low-frequency components. Furthermore, a transformative gating mechanism is incorporated into the LSTM, featuring fusion gated units to create the VT-LSTM, effectively addressing abrupt short-term information changes. Within the dual-attention network, VT-LSTM is combined with one-dimensional temporal convolution (Conv1D) to extract spatial and temporal features of data at different frequencies. This facilitates the prediction of various sub-sequences, enabling multilevel and multi-path forecasting research. The experimental evaluation uses diverse models on stock market index datasets and individual stock datasets. The results show the superior prediction accuracy of the proposed model compared to alternative methods, underscoring its practical viability in stock market forecasting applications.

DTSMLA: A dynamic task scheduling multi-level attention model for stock ranking

Predicting stock ranking is a complex and challenging task due to the intricate nature of real stock market systems. There are two main obstacles for current methods to directly using historical trading data to predict stock ranking: (1) Predicting stock rankings is a complex process influenced by a variety of factors, such as the temporal dependence of each stock, the spatial correlation between stocks, and the influence of the market on the stock; (2) The stock market’s considerable noise creates volatility in directly predicting stock rankings on a given day, presenting difficulties in training models. To overcome these two challenges, a dynamic task scheduling multi-level attention model (DTSMLA) is proposed to enhance stock ranking prediction through market index and multi-task learning. In response to complexity of stock data, we synthesize the factors affecting stocks and propose a multi-level attention prediction framework. To address the noise issue, we introduce multi-task learning and use a task scheduler algorithm to dynamically select auxiliary tasks for model training. Experimental results conducted on four real-world stock datasets have demonstrated that our method outperforms several state-of-the-art approaches.

Exploring the relationships among stocks of community wealth, state farm to school policies, and the intensity of farm to school activities

Farm to School (FTS) is a widely implemented public food procurement program in the U.S., and the number of state and Federal policies promoting it continue to grow. However, previous research has found inconclusive results associated with state-level policies. One reason may be that FTS adoption is associated with different stocks of community wealth (assets including human, social, and financial capital, net of liabilities). This research leverages the U.S. Department of Agriculture’s 2019 FTS Census, a new disaggregated database on state-level FTS policies, and a new comprehensive dataset of stocks of community wealth to assess this relationship. We find positive associations between cultural and social capital and FTS intensity (an index representing the extent of FTS participation), highlighting community assets that are often overlooked in community development programs. Further, we find that different types of state FTS policies are associated with FTS intensity; procurement policies have positive associations and education policies have negative associations. Results provide rationale for more nuanced consideration of local assets in prioritizing the types of FTS policy.

Quantitative trading models based on Sufficient Dimension Reduction and Ensemble Learning

With the continuous development of the securities market, constructing quantitative trading strategies with strong generalization ability the adaption to dynamic and changeable market environment is becoming more and more momentous in the field of quantitative finance. Under such circumstances, this paper proposes a stock prediction model based on sufficient dimension reduction theory and ensemble learning, which can be deployed on quantitative trading strategy. The proposed model on the one hand alleviates the curse of dimension by using the sufficient dimension reduction method and maximally retains variation of stock factors, on the other hand employs ensemble learning technique which can weigh model bias and variance to address the overfitting or underfitting problems. Furthermore, since ensemble learning are model free, different sub-models can be applied to improve the generalization ability of the algorithm on predicting occasions. In the section of accuracy comparison and quantitative trading strategy experiments based on real stock datasets, the proposed model demonstrates the best prediction results and adequate robustness compared to other existed methods.

Attention based adaptive spatial–temporal hypergraph convolutional networks for stock price trend prediction

Stock price trend prediction is an important and challenging issue, and accurate forecasting will effectively improve investment decisions and contribute to investment returns. Improving prediction accuracy by exploring stock correlations has received much attention in recent studies. However, there are still some issues that have not been fully considered, such as the impact of invalid correlations, low sensitivity to minor price fluctuations and dependence on priors expert information. To solve the above issues, we propose a novel spatial–temporal framework, which has several characteristics: (1) the noise-aware spatial–temporal attention that dynamically filters out invalid associations from traditional spatial attention and combines temporal attention to capture the spatial–temporal patterns of different stock series; (2) the adaptive stock hypergraph generation maps the intrinsic associations of stocks into a trainable dense matrix via adaptive node embedding; (3) the adaptive graph convolution extends the graph convolution operation from static graphs to adaptive hypergraphs for exploring the dynamic correlations. (4) Multiple stacked attention-based adaptive spatial–temporal Blocks form the end-to-end prediction framework, which uses time-aware cascaded convolution to extract fine-grained temporal features. Convincing experimental results on two stock datasets, studies on the performance on various simulation investments and the model interpretability confirm the advantages of our approach.

Adversarial unsupervised domain adaptation based on generative adversarial network for stock trend forecasting

Stock trend forecasting, which refers to the prediction of the rise and fall of the next day’s stock price, is a promising research field in financial time series forecasting, with a large quantity of well-performing algorithms and models being proposed. However, most of the studies focus on trend prediction for stocks with a large number of samples, while the trend prediction problem of newly listed stocks with only a small number of samples is neglected. In this work, we innovatively design a solution to the Small Sample Size (SSS) trend prediction problem of newly listed stocks. Traditional Machine Learning (ML) and Deep Learning (DL) techniques are based on the assumption that the available labeled samples are substantial, which is invalid for SSS trend prediction of newly listed stocks. In order to break out of this dilemma, we propose a novel Adversarial Unsupervised Domain Adaptation Network (AUDA-Net), based on Generative Adversarial Network (GAN), ad hoc for SSS stock trend forecasting. Different from the traditional domain adaptation algorithms, we employ a GAN model, which is trained on basis of the target stock dataset, to effectively solve the absence problem of available samples. Notably, AUDA-Net can reasonably and successfully transfer the knowledge learned from the source stock dataset to the newly listed stocks with only a few samples. The stock trend forecasting performance of our proposed AUDA-Net model has been verified through extensive experiments conducted on several real stock datasets of the U.S. stock market. Using stock trend forecasting as a case study, we show that the SSS forecasting results produced by AUDA-Net are favorably comparable to the state-of-the-art.

EVALUATION OF SUPPORT VECTOR MACHINE BASED STOCK PRICE PREDICTION

&#x0D; &#x0D; &#x0D; In recent years with the advent of computational power, Machine Learning has become a popular approach in financial forecasting, particularly for stock price analysis. In this paper, the authors develop a non-recurrent active trading algorithm based on stock price prediction, using Support Vector Machines on high frequency data, and compare its risk adjusted performance to the returns of a statistical portfolio predicted by the Capital Asset Pricing Model. The authors selected the three highest volume securities from a pool of 100 initially selected stock dataset to investigate the algorithmic trading strategy. The abnormal return estimates are significant and positive, and the systematic risk is lower than unity in all cases, suggesting lower risk compared to the market. Moreover, the estimated beta values for all stocks were close to zero, indicating a market independent process. The correlation analysis revealed weak correlations among the processes, supporting the potential for risk reduction and volatility mitigation through portfolio diversification. The authors tested an equally weighted portfolio of the selected three assets and demonstrated a remarkable return of 1348% during the evaluation period from July 1st, 2020, to January 1st, 2023. The results suggest that the weak form of market efficiency can be questioned, as the algorithmic trading strategy, employing a Support Vector Machine binary classification model, has consistently generated statistically significant and substantial abnormal returns using historical market data.&#x0D; &#x0D; &#x0D;

Twin extreme learning machine based on heteroskedastic Gaussian noise model and its application in short-term wind-speed forecasting

Extreme learning machine (ELM) has received increasingly more attention because of its high efficiency and ease of implementation. However, the existing ELM algorithms generally suffer from the drawbacks of noise sensitivity and poor robustness. Therefore, we combine the advantages of twin hyperplanes with the fast speed of ELM, and then introduce the characteristics of heteroscedastic Gaussian noise. In this paper, a new regressor is proposed, which is called twin extreme learning machine based on heteroskedastic Gaussian noise (TELM-HGN). In addition, the augmented Lagrange multiplier method is introduced to optimize and solve the presented model. Finally, a significant number of experiments were conducted on different data-sets including real wind-speed data, Boston housing price dataset and stock dataset. Experimental results show that the proposed algorithms not only inherits most of the merits of the original ELM, but also has more stable and reliable generalization performance and more accurate prediction results. These applications demonstrate the correctness and effectiveness of the proposed model.

An Accelerated Optimization Approach for Finding Diversified Industrial Group Stock Portfolios with Natural Group Detection

Stock portfolio optimization is always an interesting and attractive research problem due to the variety of stock markets. To find a useful stock portfolio, metaheuristic-based approaches have been presented to obtain diverse group stock portfolios (DGSPs) by considering the diversity of stock portfolios in the past. However, in the existing DGSP algorithms, two problems remain to be solved. The first is how to set a suitable group size, and the second is that the evolution process is time-consuming. To solve these problems, in this paper, an approach using grouping genetic algorithms (GGAs) was proposed for optimizing a DGSP. For setting a suitable group size, the proposed approach utilized two attributes of group stocks, including the return on equity and the price/earnings ratio. Then, to derive better stock groups, a cluster validation factor was designed, which was used as part of a fitness function. To solve the time-consumption problem, using the designed temporary chromosome, the number of stock portfolios that need to be evaluated could be reduced in the proposed approach to speed up the evolution process. Finally, experiments on two real stock datasets containing 31 and 50 stocks were conducted to show that the proposed approach was effective and efficient. The results indicated that the proposed approach could not only achieve similar returns but also accelerate the evolution process when compared with the existing algorithms.

RAdam-DA-NLSTM: A Nested LSTM-Based Time Series Prediction Method for Human–Computer Intelligent Systems

At present, time series prediction methods are widely applied for Human–Computer Intelligent Systems in various fields such as Finance, Meteorology, and Medicine. To enhance the accuracy and stability of the prediction model, this paper proposes a time series prediction method called RAdam-Dual stage Attention mechanism-Nested Long Short-Term Memory (RAdam-DA-NLSTM). First, we design a Nested LSTM (NLSTM), which adopts a new internal LSTM unit structure as the memory cell of LSTM to guide memory forgetting and memory selection. Then, we design a self-encoder network based on the Dual stage Attention mechanism (DA-NLSTM), which uses the NLSTM encoder based on the input attention mechanism, and uses the NLSTM decoder based on the time attention mechanism. Additionally, we adopt the RAdam optimizer to solve the objective function, which dynamically selects Adam and SGD optimizers according to the variance dispersion and constructs the rectifier term to fully express the adaptive momentum. Finally, we use multiple datasets, such as PM2.5 data set, stock data set, traffic data set, and biological signals, to analyze and test this method, and the experimental results show that RAdam-DA-NLSTM has higher prediction accuracy and stability compared with other traditional methods.

A knowledge graph–GCN–community detection integrated model for large-scale stock price prediction

Owing to uncertainty in the stock market, stock price prediction has always been a challenging research hotspot. In recent years, many stock prediction methods have used stock price series and technical indicators as inputs and the time series algorithm to predict, but they often ignore the influence of deeper factors such as the situation of the stock company and current situation of the stock industry. In addition, most of them predict based on small-scale stock datasets with limited characteristics and have certain defects such as bias, poor stability of prediction results, and lack of statistical significance tests on experimental results. To solve these problems, we propose a new method for stock price prediction based on knowledge graph (KG) and graph convolution neural network (GCN) models. First, stock KG is constructed, and the semantic relationships between stocks are described in the form of triples. Second, the correlations between stocks are quantified by fully utilizing their explicit/implicit relationships in the KG. Third, K-means, community detection (CD), and GCNs are merged to obtain accurate clustering results for similar stocks. Finally, the historical prices of similar stocks are used as the input characteristics of the time series models to predict stock price trends. We collect 4684 A-share market stocks in China from 2013 to 2019 and predicted the stock price trends for 762 of them. The experimental results and significance test show that the proposed method achieve the best accuracy, precision, and F1-measure on large-scale stock datasets and have the best stability, proving that the overall prediction effect outperforms that by state-of-the-art methods.

Sequential Graph Attention Learning for Predicting Dynamic Stock Trends (Student Abstract)

The stock market is characterized by a complex relationship between companies and the market. This study combines a sequential graph structure with attention mechanisms to learn global and local information within temporal time. Specifically, our proposed “GAT-AGNN” module compares model performance across multiple industries as well as within single industries. The results show that the proposed framework outperforms the state-of-the-art methods in predicting stock trends across multiple industries on Taiwan Stock datasets.

Constructing soils for climate-smart mining

Surface mining is inherently linked to climate change, but more precise monitoring of carbon dioxide (CO2) emissions is necessary. Here we combined the geolocation of mine sites and carbon stock datasets to show that if all legal active mining sites in Brazil are exploited over the next decades, 2.55 Gt of CO2 equivalent (CO2eq) will be emitted due to the loss of vegetation (0.87 Gt CO2eq) and soil (1.68 Gt CO2eq). To offset these emissions, we propose constructing soils (Technosols) from mine and other wastes for mine reclamation. We show that this strategy could potentially offset up to 60% (1.00 Gt CO2eq) of soil-related CO2 emissions. When constructed with suitable parent materials, Technosols can also restore important soil-related ecosystem services while improving waste management. The construction of healthy Technosols stands out as a promising nature-based solution towards carbon-neutral mining and should, therefore, be considered in future environmental policies of major mining countries.

Portfolio Optimization Based on Return Prediction and Semi Absolute Deviation (SAD)

A portfolio is a collection of investment financial assets managed by financial institutions or individuals. In investment activities, investors expect minimal loss risk and optimal stock portfolio weight to get maximum profit. Investors can monitor changes in stock index values to compare portfolio performance. This research has discussed how to build a portfolio based on stock datasets with the LQ45 index using return predictions from the artificial neural network (ANN) method with semi-absolute deviation (SAD). Furthermore, the portfolio is optimized by looking for weights that match it. After that, a comparison of portfolio performance was carried out using the Sharpe ratio (SR) method between the semi-absolute deviation (SAD) portfolio and the portfolio resulting from the formation of the equal weight (EW) portfolio. Portfolio performance with ANN prediction and SAD is better than equal-weight portfolios in terms of mean return, standard deviation, and sharpe ratio for portfolios with few stocks, namely 2 and 3 stocks. In addition, a portfolio with a higher number of stocks can make the portfolio value from the ANN close prediction algorithm process and the selection of weights based on SAD is better than portfolios with equal weight for each list of stocks in the portfolio.

A Graph Neural Network Node Classification Application Model with Enhanced Node Association

This study combines the present stage of the node classification problem with the fact that there is frequent noise in the graph structure of the graph convolution calculation, which can lead to the omission of some of the actual edge relations between nodes and the appearance of numerous isolated nodes. In this paper, we propose the graph neural network model ENode-GAT for improving the accuracy of small sample node classification using the method of external referencing of similar word nodes, combined with Graph Convolutional Neural Network (GCN), Graph Attention Network (GAT), and the early stop algorithm. In order to demonstrate the applicability of the model, this paper employs two distinct types of node datasets for its investigations. The first is the Cora dataset, which is widely used in node classification at this time, and the second is a small-sample Stock dataset created by Eastern Fortune’s stock prospectus of the Science and Technology Board (STB). The experimental results demonstrate that the ENode-GAT model proposed in this paper obtains 85.1% classification accuracy on the Cora dataset and 85.3% classification accuracy on the Stock dataset, with certain classification advantages. It also verifies the future applicability of the model to the fields of stock classification, tender document classification, news classification, and government announcement classification, among others.

APSO-TA-LSTM: a long and short term memory model combining time attention and adaptive particle swarm optimization for stock forecasting

A new stock forecasting model that combines time attention and adaptive particle swarm optimization with LSTM (APSO-TA-LSTM) is proposed to improve the forecasting ability of neural networks for financial time series. The model uses a two-layer LSTM network to encode stock information within the time window and employs time attention to strategically focus on dependencies among time series features for more accurate feature representations. Additionally, the proposed adaptive particle swarm optimization algorithm is used to pick out the key parameters of the network structure and enhance the overall prediction performance. Finally, the experimental results on three stock datasets validate the innovation and effectiveness of our method, and this work will have a broad application prospect in the study of financial time series.

Techniques for Stock Market Prediction: A Review

Stock market forecasting has long been viewed as a vital real-life topic in economics world. There are many challenges in stock market prediction systems such as the Efficient Market Hypothesis (EMH), Nonlinearity, complex, diverse datasets, and parameter optimization. A stock's value on the stock market fluctuates due to many factors like previous trends of the stock, the current news, twitter feeds, any online customer feedbacks etc. In this paper, the literature is critically analysed on approaches used for stock market prediction in terms of stock datasets, features used, evaluation metrics used, statistical, machine learning and deep learning techniques along with the directions for the future. The focus of this review is on trend and value prediction for stocks. Overall, 68 research papers have been considered for review from years 1998-2023. From the review, Indian stock market datasets are found to be most frequently used datasets. Evaluation metrics used commonly are accuracy and Mean Absolute Percentage Error. ARIMA is reported as the most used frequently statistical technique for stick market prediction. Long-Short Term Memory and Support Vector Machine are the commonly used algorithms in stock market prediction. The advantages and disadvantages of frequently used evaluation metrics, machine learning, deep learning and statistical approaches are also included in this survey.