Fig 3 - uploaded by Binod Rimal
Content may be subject to copyright.
Source publication
The rapid advancement in artificial intelligence and machine learning techniques, availability of large-scale data, and increased computational capabilities of the machine opens the door to develop sophisticated methods in predicting stock price. In the meantime, easy access to investment opportunities has made the stock market more complex and vol...
Contexts in source publication
Context 1
... í µí±¤ denotes all learnable parameters. Since the final hidden state ℎ í µí± encodes the most information from the input sequence, it is converted to a vector using a dense layer. Fig. 3 represents the proposed model framework. The input sequence is created using í µí± = 11 features with time step m, as shown in top part of the figure. Then at time í µí±¡ − 1, the input í µí± í µí±¡−1 , a matrix of size 11 × í µí±, together with ℎ í µí±¡−2 and í µí± í µí±¡−2 is fed into the LSTM. For the next step, the output ℎ í ...
Context 2
... LSTM. This process continues until the final input sequence í µí± í µí± with corresponding output ℎ í µí± , a vector of length equal with given number of neurons of the last LSTM layer. Finally, ℎ í µí± is transmitted to a fully connected layer where linear activation function is used to predict the closing price as shown in the last part of Fig. 3. ...
Similar publications
With the rapid development in technology and digitization of financial market in recent years, an increasing number of people have begun to invest in the stock market. Accurately anticipating the change in stock price can reduce stock investors’ investment risk while also increasing their investment return. Stock price prediction is frequently a no...
Stock price prediction is a prevalent research field in both industry and academia. There is a pressing demand to develop a prediction model that captures the pattern of the financial activities with high precision to make an informed decision. Stock price prediction is challenging due to the complex, incomplete, fuzzy, nonlinear, and volatile natu...
The goal of the work is to enhance existing financial market forecasting frameworks by including an additional factor–in this example, a collection of carefully chosen tweets—into a long-short repetitive neural channel. In order to produce attributes for such a forecast, this research used a unique attitude analysis approach that combined psycholog...
The decline and increase in the price of shares of plantation companies is a problem for investors in making decisions to buy or sell shares. Factors influencing the movement of plantation stock prices include CPO commodity price fluctuations, world oil price fluctuations, Rupiah exchange rate fluctuations, government regulations and policies, dema...
Predicting the stock market remains a difficult field because of its inherent volatility. With the development of artificial intelligence, research using deep learning for stock price prediction is increasing, but the importance of applying a prediction system consisting of preparing verified data and selecting an optimal feature set is lacking. Ac...
Citations
... In finance neural networks are used in the study of financial series (Krollner et al., 2010;Zhang et al., 2023), stock prices (Bodart & Candelon, 2009;Niu et al., 2023), stock market indices (Radomska, 2021;Alkhoshi & Belkasim, 2018;Kumar & Murugan, 2013;Moghaddam et al., 2016;Song & Choi, 2023;Bhandari et al., 2022;Al-Akashi, 2022), forecasting volatility of many financial variables (Donaldson & Kamstra, 1996a, b;Salchenberger et al., 1992;Kristjanpoller et al., 2014;Ramos-Pérez et al., 2019;Liu et al., 2017;Hamid & Iqbal, 2004;Sahiner et al., 2021). For example, volatility analysis of the S&P 500 index using the LTSM40 network was conducted (Xiong et al., 2016). ...
The paper investigates the issue of volatility of stock index returns on the Warsaw Stock Exchange (WIG20 index returns volatility). The purpose of this review is to compare how other stock market indexes as HANG SENG, NIKKEI 225, FTSE 250, DAX, S&P 500 and NASDAQ 100 influance the volatility of WIG20 index returns. The innovation of this work is the usage of a new neural network with three different activation functions to predict future volatility of WIG20 index returns. The input for this network is the last 3 values of WIG20 index returns volatility and the last 3 values of one of the considered foreign index returns volatility. As measurements for the best forecasting performance of neural networks are taken common used forecast errors: ME (mean error), MPE (mean percentage error), MAE (mean absolute error), MAPE (mean absolute percentage error), RMSE (root mean square error). The study shows that the Polish stock market is mainly influenced by the European and US markets.
... However, selecting the right stocks at the right time remains a daunting challenge due to various factors and circumstances influencing stock prices. The inherently nonlinear, volatile, fuzzy, noisy, nonparametric, and deterministic chaotic nature of stock behavior (Ahangar et al. 2010;Bhandari et al. 2022;Gandhmal and Kumar 2019;Weng et al. 2017) adds complexity. Utilizing reliable sources of public sentiment regarding stock prices can enhance investment decisions by employing stock price prediction models, thereby aiding in mitigating market uncertainty. ...
... ANNs are an excellent model for predicting stock price movement direction due to their ability to learn and model complex, nonlinear relationships within financial data (Bhandari et al. 2022;Dahal et al. 2023;Pokhrel et al. 2022Pokhrel et al. , 2024. ANNs, inspired by the human brain, consist of interconnected layers of nodes that can automatically detect intricate patterns and interactions in the data. ...
... ANNs are an excellent model for predicting stock price movement direction due to their ability to learn and model complex, nonlinear relationships within financial data (Bhandari et al. 2022;Dahal et al. 2023;Pokhrel et al. 2022Pokhrel et al. , 2024. ANNs, inspired by the human brain, consist of interconnected layers of nodes that can automatically detect intricate patterns and interactions in the data. ...
Predicting stock market movement direction is a challenging task due to its fuzzy, chaotic, volatile, nonlinear, and complex nature. However, with advancements in artificial intelligence, abundant data availability, and improved computational capabilities, creating robust models capable of accurately predicting stock market movement is now feasible. This study aims to construct a predictive model using news headlines to predict stock market movement direction. It conducts a comparative analysis of five supervised classification machine learning algorithms—logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), and artificial neural network (ANN)—to predict the next day’s movement direction of the close price of the Nepal Stock Exchange (NEPSE) index. Sentiment scores from news headlines are computed using the Valence Aware Dictionary for Sentiment Reasoning (VADER) and TextBlob sentiment analyzer. The models’ performance is evaluated based on sensitivity, specificity, accuracy, and the area under the receiver operating characteristic (ROC) curve (AUC). Experimental results reveal that all five models perform equally well when using sentiment scores from the TextBlob analyzer. Similarly, all models exhibit almost identical performance when using sentiment scores from the VADER analyzer, except for minor variations in AUC in SVM vs. LR and SVM vs. ANN. Moreover, models perform relatively better when using sentiment scores from the TextBlob analyzer compared to the VADER analyzer. These findings are further validated through statistical tests.
... The second set of variables demonstrated in Table 1 is the technical indicators, including moving averages (10 days, 20 days, and 50 days), moving average convergence divergence (MACD), average true range (ATR), relative strength index (RSI), and money flow index (MFI). The technical indicators' detailed explanation is included in the research articles; [41,42]. Active traders extensively use them in the market as they are primarily designed to analyze short-term price movements [43][44][45][46][47][48][49][50]. ...
... Information is selectively disseminated across the architecture with the help of a series of sigmoid layers, hyperbolic tangent layers, and the pointwise multiplication operation. The detailed structure of the LSTM unit is explained in [41]. ...
Real estate significantly contributes to the broader stock market and garners substantial attention from individual households to the overall country’s economy. Predicting real estate trends holds great importance for investors, policymakers, and stakeholders to make informed decisions. However, accurate forecasting remains challenging due to it’s complex, volatile, and nonlinear behavior. This study develops a unified computational framework for implementing state-of-the-art deep learning model architectures the long short-term memory (LSTM), the gated recurrent unit (GRU), the convolutional neural network (CNN), their variants, and hybridizations, to predict the next day’s closing price of the real estate index S &P500-60. We incorporate diverse data sources by integrating real estate-specific indicators on top of fundamental data, macroeconomic factors, and technical indicators, capturing multifaceted features. Several models with varying degrees of complexity are constructed using different architectures and configurations. Model performance is evaluated using standard regression metrics, and statistical analysis is employed for model selection and validation to ensure robustness. The experimental results illustrate that the base GRU model, followed by the bidirectional GRU model, offers a superior fit with high accuracy in predicting the closing price of the index. We additionally tested the constructed models on the Vanguard Real Estate Index Fund ETF and the Dow Jones U.S. Real Estate Index for robustness and obtained consistent outcomes. The proposed framework can easily be generalized to model sequential data in various other domains.
... The results showed that the proposed model outperforms the baseline models and stock trading strategies. Bhandari et al. (2022) used an LSTM neural network to predict the S &P 500 index combining fundamental market data and technical indicators. The results showed that the performance of a single-layer model is found to overcome the prediction accuracy of the multilayer models, concluding that the LSTM model is a promising tool for stock price prediction. ...
Recent advances in Machine Learning and, especially, Deep Learning, have led to applications of these areas in different fields of knowledge, with great emphasis on stock market prediction. There are two main approaches in the literature to predict future prices in the stock market: (1) considering historical stock prices; and (2) considering news or social media documents. Despite the recent efforts to combine these two approaches, the literature lacks works in which both strategies are performed with Deep Learning, which has led to state-of-art results in many regression and classification tasks. To overcome these limitations, in this work we proposed a new Deep Learning-based approach to predict the Brazilian stock market combining the use of historical stock prices, financial technical indicators, and financial news. The experiments were performed considering the period from 2010 to 2019 with the Ibovespa index and the historical prices of the following Brazilian companies: Banco do Brasil, Itaú, Ambev, and Gerdau, which have significant contribution to the Ibovespa index. Our results show that the combination of stock prices, technical indicators and news improves the stock market prediction considering both the prediction error and return-of-investment.
... For the final models (Table 3), the author retained all variables in the model for which the p-value was found to be 0.05 or less. The author also reports the coefficient of determination (R 2 ), which estimates the proportion of variation in the outcome variable that is explained by the statistical model (Bhandari et al., 2022;Rights & Cole, 2018;Turney, 2022), and the frequency with which each variable or factor appeared in the different models in the analysis. Coefficients of determination in psychological studies and related fields in human behavior usually have R 2 values which are consistently below .5 (Meyer et al., 2001;Owens et al., 2021;Hayes et al., 2022). ...
... Using the experience in similar studies and suggestions in the literature, the author adopted the following rule of thumb for analyzing R 2 values: R 2 values ranging from above 0.00-0.3, 0.3-0.5, and 0.5 or more are respectively considered as small, medium, and large with regard to predictive ability of the model (Bhandari et al., 2022;Jiang & Smith, 2002). ...
This study explores the association between youth antisocial behavior outcomes and predictors including individual factors (youth attitudes) and interpersonal factors (friends and parents) in a low income, immigrant, and ethnic community in Pomona, California USA. Social cognitive theory is used in a cross sectional design to perform correlation and multiple regression analysis on survey data from 2,693 high school students. While factors related to youths and their best friends were associated with youth antisocial behavior outcomes, parental factors did not show any association with youth antisocial behavior outcomes in both correlation and regression analysis. These results conflict with dominant trends in the literature and underscore the need for at risk communities such as Pomona to develop their own community data so that intervention programs can respond to their unique circumstances. The following intervention strategies are recommended for youth program designers in Pomona: prioritize individual youth and friends in intervention activities; develop intervention strategies to grow parental influence on youth antisocial behavior such as integrating engagement tools which immigrant, minority, and low income parents could use to counter antisocial risk factors and enhance protective factors. Future research should investigate the factors that drive the influence of best friends on youth antisocial behavior particularly how these trump parental influence in Pomona and other disadvantaged environments. The research findings should shape the intervention strategies recommended above.
... Due to the approximately 2500 data groups (O, H, L, C prices) in each market over 10 years period, and based on the findings of Bhandari et al. (2022) that a single hidden layer LSTM model outperformed models with multiple hidden layers in terms of stock market predictive performance, this paper employed a single-layer LSTM model with 50 neurons. Considering that price differences are crucial characteristics in the stock market and prices cannot be negative, the ReLU (Rectified Linear Unit) activation function was chosen. ...
This study examines the influence of the sliding window in the LSTM model on its predictive performance in the stock market. The investigation encompasses three aspects: the impact of the stationarity of the original data, the effect of the time interval, and the influence of the input order of data. Additionally, a standard VAR model is established for a comparative benchmark. The experimental dataset comprises the daily stock index prices of the six major stock markets from the January 2010 to December 2019. The experimental results demonstrate that stationary input data enhances the predictive performance of the LSTM model. Furthermore, shorter time interval tends to yield improved outcomes, while the order of input data does not impact the performance of the LSTM. Although the predictive capability of the LSTM model may not consistently surpass that of the standard VAR model, which is different from the previous research, it serves to compensate for the conditional limitations associated with VAR model construction.
... In recent years, many researchers have explored this topic from various perspectives. They have employed different approaches such as statistical methods [3,7,9,15], supervised learning methods, with Long Short-Term Memory (LSTM) being a typical example [4,8,14,16], and unsupervised learning [13]. Statistical methods require several factors to achieve good results, such as data reliability and adherence to standards and assumptions. ...
... In order to better verify the superiority of the QCRGRNN algorithm, the RNN model and its variant network LSTM model were established for comparison. Four models of RNN [31], LSTM [32], GRU [33] and QCRGRNN were selected to predict methane production. The prediction results are shown in figure 13. ...
Methane as a renewable energy source has become a hot topic in recent years. Methane is a bioenergy source produced during the anaerobic digestion of organic waste, and the anaerobic digestion process must be monitored and controlled to produce the required amount of methane in a stable manner. Mathematical modeling is used to simulate digester operation to predict the biogas production from anaerobic digestion, to avoid reactor loading or performance degradation, and to ensure efficient operation of the system. In this paper, a Quantum Convolutional Reconstruction Gated Recurrent Neural Network is proposed. The original data features are extracted by convolutional neural network to reduce the dimensionality and retain the information, the parameterized quantum circuit is integrated in the gating recurrent unit, and the quantum reset gate and quantum update gate are constructed. The information extracted by the Convolution Neural networks is input into the quantum gated recurrent neural network, and the quantum storage unit integrates the information into the hidden layer state, thus processing the hidden layer state information more efficiently. The experimental results show that the prediction accuracy of the A Quantum Convolution Reconstructed Gated Recurrent Neural Network is improved from 81.95 to 88.21%, and the MAE value is reduced from 54.53% to 37.38%.
... In recent years, many researchers have explored this topic from various perspectives. They have employed different approaches such as statistical methods [3,7,9,15], supervised learning methods, with Long Short-Term Memory (LSTM) being a typical example [4,8,14,16], and unsupervised learning [13]. Statistical methods require several factors to achieve good results, such as data reliability and adherence to standards and assumptions. ...
Stock market prediction holds significant importance in the world of finance, captivating the attention of both investors and financial researchers. The integration of artificial intelligence and advancements in computational power has led to substantial improvements in predicting stock prices, surpassing the effectiveness of traditional programmed prediction methods. In this paper, we explore three distinct and innovative methods for stock price prediction: Long Short-Term Memory (LSTM), LSTM combined with Simple Moving Average (LSTM-SMA), and LSTM combined with Exponential Moving Average (LSTM-EMA). Our analysis is conducted using a comprehensive historical dataset of Apple's stock prices, and the performance of each model is rigorously evaluated using critical metrics, including Mean Absolute Percentage Error (MAPE), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R2 score. Additionally, the training time for each model is taken into account. The results show that all three models LSTM, LSTM-SMA, and LSTM-EMA give good prediction results for Apple's stock price, in which the LSTM model gives the best prediction results for the 21-day cluster. However, in terms of computational time, the LSTM-SMA model and LSTM-EMA model are more efficient than the LSTM model. These findings highlight the potential of integrating advanced techniques to achieve more accurate and efficient stock price predictions.
... Numerous studies have used machine learning models to predict or forecast traditional stock indices. In recent years, most studies have used machine learning techniques to analyse digital assets price predictions (Bhandari et al., 2022;Yan, 2023). However, few studies have investigated token assets closing price forecasting, which can use machine learning and deep learning for analysis (Branny et al., 2022;Ghosh et al., 2023). ...
... This study is used for feature selection with the help of Sklearn-wrapped Adaboost regressors (Drucker, 1997;Freund and Schapire, 1997). In recent studies, also use long shortterm memory (LSTM) algorithm for prediction various assets price, returns and volatility (Bhandari et al., 2022;Chhajer et al., 2022;Fischer and Krauss, 2018;Peng et al., 2018;Yan, 2023). A LSTM algorithm combined with the dropout layer and some basic machine learning algorithms is constricted to predict AI-based tokens' price futures. ...
