Improving Stock Price Prediction with GAN-Based Data Augmentation

Stock price prediction is useful for investors to see how the prospects of a company's stock investment in the future. Stock price prediction can be used to anticipate the deviation of stock prices. It can also helps investors in decision making. Artificial Neural Networks do not require mathematical models but data from problems to be solved. Information is conveyed through the data, and the Artificial Neural Network filters the information through training. Therefore, Artificial Neural Network is appropriate to solve the problem of stock price prediction. Learning method that will be used to predict stock price is Supervised Learning with Backpropagation algorithm. With this algorithm, networks can be trained using stock price data from the previous time, classify it and adjust network link weight as new input and forecast future stock prices. By using ANN, time series prediction is more accurate. After analyzing the problem of stock price movement system, the writer can know the pattern of what variables will be taken for further insert into the stock price forecasting system. This application can be used for stock price forecasting technique, so it will be useful for beginner investor as well as advanced investor as reference to invest in capital market. Implementing supervised learning backpropagation method will get accurate forecasting results more than 98%.Keyword - artificial neural network, stock, backpropagation.

In the era of artificial intelligence, machine learning methods are successfully used in various fields. Machine learning has attracted extensive attention from investors in the financial market, especially in stock price prediction. However, one argument for the machine learning methods used in stock price prediction is that they are black-box models which are difficult to interpret. In this paper, we focus on the future stock price prediction with the historical stock price by machine learning and deep learning methods, such as support vector machine (SVM), random forest (RF), Bayesian classifier (BC), decision tree (DT), multilayer perceptron (MLP), convolutional neural network (CNN), bi-directional long-short term memory (BiLSTM), the embedded CNN, and the embedded BiLSTM. Firstly, we manually design several financial time series where the future price correlates with the historical stock prices in pre-designed modes, namely the curve-shape-feature (CSF) and the non-curve-shape-feature (NCSF) modes. In the CSF mode, the future prices can be extracted from the curve shapes of the historical stock prices. Conversely, in the NCSF mode, they can’t. Secondly, we apply various algorithms to those pre-designed and real financial time series. We find that the existing machine learning and deep learning algorithms fail in stock price prediction because in the real financial time series, less information of future prices is contained in the CSF mode, and perhaps more information is contained in the NCSF. Various machine learning and deep learning algorithms are good at handling the CSF in historical data, which are successfully applied in image recognition and natural language processing. However, they are inappropriate for stock price prediction on account of the NCSF. Therefore, accurate stock price prediction is the key to successful investment, and new machine learning algorithms handling the NCSF series are needed.

Prediction of stock prices has been the primary objective of an investor. Any future decision taken by the investor directly depends on the stock prices associated with a company. This work presents a hybrid approach for the prediction of intra-day stock prices by considering both time-series and sentiment analysis. Furthermore, it focuses on long short-term memory (LSTM) architecture for the time-series analysis of stock prices and Valence Aware Dictionary and sEntiment Reasoner (VADER) for sentiment analysis. LSTM is a modified recurrent neural network (RNN) architecture. It is efficient at extracting patterns over sequential time-series data, where the data spans over long sequences and also overcomes the gradient vanishing problem of RNN. VADER is a lexicon and rule-based sentiment analysis tool attuned to sentiments expressed in social media and news articles. The results of both techniques are combined to forecast the intra-day stock movement and hence the model named as LSTM-VDR. The model is first of its kind, a combination of LSTM and VADER to predict stock prices. The dataset contains closing prices of the stock and recent news articles combined from various online sources. This approach, when applied on the stock prices of Bombay Stock Exchange (BSE) listed companies, has shown improvements in comparison to prior studies.

In Korea, because of the high interest in stock investment, many researchers have attempted to predict stock prices using deep learning. Studies to predict stock prices have been continuously conducted. However, the type of stock data that is suitable for deep learning has not been established, and it has not been confirmed that the developed stock prediction model can actually result in a profit. To date, designing a good deep learning model depends on how well the user can extract the features that represent all the characteristics of the training data. Among the various available features for training and test data, we determined that the use of event binary features can make stock price prediction models perform better. An event binary feature refers to a 0 or 1 value describing whether an indicator is satisfied (1) or not (0) for any given day and stock. We proposed and compared a stock price prediction model with three different feature combinations to verify the importance of binary features. As a result, we derived a prediction model that defeated the market (KOSPI and KODAQ (KOSPI (Korea Composite Stock Price Index) and KOSDAQ (Korean Securities Dealers Automated Quotations) is Korean stock indices)). The results suggest that deep learning is suitable for stock price prediction.

From past to present, the prediction of stock price in stock market has been a knotty problem. Many researchers have made various attempts and studies to predict stock prices. The prediction of stock price in stock market has been of concern to researchers in many disciplines, including economics, mathematics, physics, and computer science. This study intends to learn fluctuation of stock prices in stock market by using recently spotlighted techniques of deep learning to predict future stock price. In previous studies, we have used price-based input-features to measure performance changes in deep learning models. Results of this studies have revealed that the performance of stock price models would change according to varied input-features configured based on stock price. Therefore, we have concluded that more novel input-feature in deep learning model is needed to predict patterns of stock price fluctuation more precisely. In this paper, for predicting stock price fluctuation, we design deep learning model using 715 novel input-features configured on the basis of technical analyses. The performance of the prediction model was then compared to another model that employed simple price-based input-features. Also, rather than taking randomly collected set of stocks, stocks of a similar pattern of price fluctuation were filtered to identify the influence of filtering technique on the deep learning model. Finally, we compared and analyzed the performances of several models using different configuration of input-features and target-vectors.

In recent years, the development of technology and artificial intelligence has brought forth new opportunities in analyzing and predicting stock prices. One of the approaches used is the Neural Network algorithm, which is a part of the branch of artificial intelligence known as Deep Learning. This algorithm can learn complex patterns and relationships among data by modeling inspired by the human neural network. This research utilizes the Neural Network for stock price prediction and aims to understand the application of Neural Network in predicting stock prices, which can benefit investors and market participants. Additionally, historical stock price data can be used as input for the Neural Network algorithm. The Neural Network is a frequently used algorithm for accurate predictions and is widely employed in prediction-based or forecasting research. The result of this research is the Root Mean Squared Error (RMSE) value of 19.734 +/- 0.000. The use of the Neural Network as an algorithm for stock price prediction provides investors with valuable information for making investment decisions for companies..

This research paper aims to use the deep learning model Long Short-Term Memory (LSTM) for the stock prediction model under the financial crisis of COVID-19. The financial impact of the COVID-19 has brought many of the world's indexes down. The impact of the financial crisis is even riskier for an emerging country such as Indonesia where foreign investors tend to take out their investments in emerging countries in financial crisis events. The application of deep learning in financial time series applications such as stock price prediction has been researched extensively. This study used the (Bidirectional LSTM) BiLSTM model which is a variation of the LSTM model to predict stock closing price. The stock prediction is applied to a selected company from the Indonesian stock market using historical prices. The model is then evaluated using metrics Mean Absolute Percentage Error (MAPE) and Symmetric Mean Absolute Percentage Error (SMAPE). A graphical comparison between the actual price and predicted price of the stock is charted to study the stock price movement. To study the impact during COVID-19 on the stock prices, an intervention analysis is conducted along with the Wilcoxon model. The stock price prediction model can forecast the price of stocks before and during the financial crisis with minimal error. The intervention analysis result showed that health sectors have a positive effect while other sectors such as transportation, finance, information technology, and entertainment have a negative effect during the financial crisis of COVID-19. Being able to analyze and study the stock price movement of stocks is beneficial to investors in understanding the impact of the financial crisis on some industries and the behavior of certain stocks or industries under the circumstances which can lead to alternate investment strategies and decision making.

Recently, in the building and infrastructure fields, studies on defect detection methods using deep learning have been widely implemented. For robust automatic recognition of defects in buildings, a sufficiently large training dataset is required for the target defects. However, it is challenging to collect sufficient data from degrading building structures. To address the data shortage and imbalance problem, in this study, a data augmentation method was developed using a generative adversarial network (GAN). To confirm the effect of data augmentation in the defect dataset of old structures, two scenarios were compared and experiments were conducted. As a result, in the models that applied the GAN-based data augmentation experimentally, the average performance increased by approximately 0.16 compared to the model trained using a small dataset. Based on the results of the experiments, the GAN-based data augmentation strategy is expected to be a reliable alternative to complement defect datasets with an unbalanced number of objects.

The maturity of 5G technology provides a guarantee for increasingly large communication networks, while the resources required for communication and computation are also increasing, and reasonable resource allocation can improve the efficiency of network communication and reduce the consumption of communication resources. Existing deep learning methods have been able to predict network traffic to a certain extent, so as to solve the communication efficiency and resource consumption problems in the field of integrated sensing, communication and computation (ISCC) through rational resource allocation. However, the following problems still exist: (1) The feature learning ability of the prediction model is insufficient, and the prediction accuracy needs to be improved. (2) Powerful and complex deep learning methods lead to an increase in the prediction cost of the model. To address these problems, this paper proposes a deep learning method based on a lightweight hybrid attention network. In order to capture the key features of 5G data more effectively, an efficient hybrid attention mechanism (EHA) is proposed. After this attention is applied to convolution, the key information can be well enhanced. We use depthwise separable convolution in feature extraction, which greatly improves the efficiency of lightweight convolution layer (LC) in feature extraction. Combined with the efficient hybrid attention mechanism (EHA), the proposed model has better lightweight properties. Experimental results show that the model proposed in this paper has lower RMSE and MAE values on the three datasets, as well as fewer parameters and computational effort compared to the baseline scheme.

This paper proposes a stock price prediction model, which extracts features from time series data, news, and comments on the news, for prediction of stock price and evaluates its performance. In this research, we do not take account of text contents of news and user comments, but just consider numerical features of news and communication dynamics appeared in comments on the Web as well as historical time series data. We model the stock price movements as a function of these input features and solve it as a regression problem in a Multiple Kernel Learning regression framework. Experimental results show that our proposed method consistently outperforms other baseline methods in terms of magnitude prediction measures such as MAE, MAPE and RMSE for three companies' stocks. They specifically show that the features other than stock prices themselves improved the performance.

This study aims to compare the effectiveness of two deep learning models, namely Convolutional Neural Network (CNN) and combined CNN with Gated Recurrent Unit (CNN-GRU), in predicting Petrobras stock prices. Using historical stock price data from Yahoo Finance for the period 2017-2023, this study evaluates the performance of both models based on several evaluation metrics. The results showed that the CNN-GRU model outperformed the pure CNN model in all evaluation metrics, with an increase in RMSE value of 4.17% and an increase in R² value of 0.47% on the test data. The CNN-GRU model achieved 96.14% accuracy on the test data, while the CNN model achieved 96.04%. These findings indicate that the integration of CNN's feature extraction capabilities with GRU's temporal dependency modeling capabilities can improve stock price prediction accuracy. This research contributes to the computational finance literature by presenting an in-depth comparative analysis of the application of hybrid deep learning architectures in stock market prediction.

The accurate prediction of subway passenger flow is of great significance for the intelligent management of urban rail transit systems. However, traditional static network prediction models struggle to capture the dynamic correlations of subway passenger flow, and the graph structure generated by the DTW algorithm may overlook the differences between passenger flows. Therefore, this paper proposes a GAT-LSTM-Attention prediction model based on a graph structure constructed from the differences in passenger flow. Firstly, a graph structure is constructed based on the differences in peak and valley values between stations, so that the prediction of a station is not affected by stations that exhibit significant differences. Secondly, the periodic characteristics and weather features that may affect the passenger flow prediction are extracted and fused. Finally, the prediction is carried out by the GAT-LSTM-Attention. Through the dual attention mechanism of GAT and Attention, the model is capable of effectively capturing spatial features and key information. The experimental results show that the proposed model has lower MAE and RMSE error values compared with the baseline models. In the 30-minute and 60-minute prediction of outbound and inbound passenger flow, the proposed model achieves the higher R² values of 0.99, 0.99, 0.98, and 0.97. And the proposed graph structure based on passenger flow difference has lower MAE and RMSE error values than conventional methods for constructing graphical structures. This study provides a valuable method for passenger flow prediction in urban rail transit systems.

Area under the curve may hide poor generalisation to external datasets

In the context of the rapid development of the global electric vehicle industry, Tesla's stock price is significantly affected by factors such as policies and market fluctuations. In order to improve the accuracy of its stock price prediction, this study proposes a hybrid model based on Monte Carlo simulation and LSTM neural network, introduces the policy pressure index to quantify the impact of China's "three guarantees" and other policies, and designs a dynamic weight adjustment algorithm. The experiment uses Tesla's stock price data from January 2016 to December 2023, with mean square error (MSE), mean absolute error (MAE) and determination coefficient (R²) as indicators. The results show that the MSE of the hybrid model is 18.72, which is 23.5% lower than that of the single Monte Carlo simulation and 19.2% lower than that of the single LSTM model; MAE is 4.21, which is 21.8% lower than the former and 17.6% lower than the latter; R² reaches 0.912. During the implementation of China's "Three Guarantees" policy in January 2022, the model prediction error was reduced by 35.7% compared with the traditional method, and the prediction accuracy was improved by 15.3% when the market volatility exceeded 30%. The experiment verified the effectiveness of the model in high volatility scenarios, providing investors with more reliable decision-making support.

Machine learning has been increasingly used in stock price prediction with outstanding success. Decision tree regression models and linear regression models are both important models for predicting stock prices. The paper use decision tree regression and linear regression models to predict the opening price, closing price, high price and low price of Apple's stock price data respectively. The prediction effects of the two models are evaluated by the indicators of goodness of fit, mean square error, root mean square error and mean absolute error, and the prediction effects of the two models are compared. This experimental concludes that the decision tree regression model has better and more advantageous prediction results compared to the linear regression model. This study has guiding significance for machine learning in predicting stock prices when choosing a basic model or a combination of models for prediction.