Cristiane Orquisa Fantin, Eli Hadad
Mackenzie Presbyterian University, São Paulo, Brazil.
DOI: 10.4236/jcc.2022.1012003   PDF    HTML   XML   555 Downloads   2,618 Views   Citations

Abstract

This study maps the academic literature on Stock Price Forecasting with Long-Term Memory Artificial Neural Networks—RNA LSTM. The objective is to know if it is suitable for time series studies, especially for stock price projection. Through bibliometric analysis and systematic literature review, it is observed that 333 authors wrote on the topic between 2018 and March 2022, and the journals Expert Systems with Applications, IEEE Access, Big Data Journal and Neural Computing and Applications, published the most relevant articles. Of the 99 articles published in this period, 43 are associated with Chinese institutions, the most cited being that of Kim and Won, who studies the volatility of returns and the market capitalization of South Korean stocks. The basis of 65% of the studies is the comparison between the RNN LSTM and other artificial neural networks. The daily closing price of shares is the most analyzed type of data, and the American (21%) and Chinese (20%) stock exchanges are the most studied. 57% of the studies include improvements to existing neural network models and 42% new projection models.

Keywords

Stock Price Forecasting, Long-Term Memory, Backpropagation, Bibliometric Analysis, Systematic Review

Share and Cite:

1. Introduction

The financial market is characterized by being a dynamic, complex and non-linear system, characterized by data intensity, noise, non-stationary nature, unstructured and with a high degree of uncertainty [1]. As so many factors interact simultaneously, such as political events, macro and microeconomic conditions and investor expectations, predicting these movements is a very challenging task.

The growing role that the stock market plays in the world economy stimulates the development of research aimed at building theories, involving the topic of stock price prediction, and accurate methods are crucial for the management of investor portfolios.

Assessing expected returns relative to total exposure assumes that portfolio managers understand the distribution of the portfolio. Specialists can model the influence of tangible assets in relation to market value, but not of intangible assets such as rights, experiences or brand equity.

An important ally in the quest to minimize risk in relation to exposure, artificial intelligence and machine learning with their neural networks have provided a great balance of quality in recent decades, improving detection, diagnosis, prediction and problem solving [2]; this is because in this market, future events are at least partially dependent on past events and data [3], and not entirely random.

The aim of this study is to map and analyze the published academic literature on stock price prediction using artificial neural networks Long-Term Memory Artificial Neural Networks—RNA LSTM. To this end, a bibliometric analysis and a systematic review of the literature on the subject are carried out, during the period from January 1, 2000 to March 31, 2022, with a final sample of 99 articles. Bibliometric analysis refers to quantitative analysis, which is developed by counting frequencies and co-citations. The systematic review, a qualitative analysis, considers the correlation between the most significant themes, but still little studied by the academy. The research base used comes from the Web of Science—WoS database, and both the bibliometric analysis and the systematic review do not dispense with the use of R, RStudio, Biblioshiny and VOSViewer software. In the bibliometric analysis, the verification of the main laws is adopted: Lotka [4] and Bradford [5].

The literature review is presented in item 2, with the identification of theories and methods of forecasting stock prices with multilayer perceptron artificial neural networks mentioned in the articles of the final sample. The bibliometric analysis and systematic review methodologies are described in item 3, and in item 4 the results of both methodologies are reported, with descriptive statistics of the most relevant characteristics of the articles in the final sample and the knowledge gaps on the topic. Item 5 presents the conclusions, paths for future studies and limitations of this research.

2. Literature Review

The current stock price of a publicly traded company reflects the company’s past operation, current timing and future profitability prospects. To obtain a more accurate projection of the stock price, several types of studies have already been carried out. The most classic ones focus on the financial data of the target companies, added to micro and macroeconomic aspects. However, the non-linear and non-stationary features of financial data sequences make predictions more challenging [6]. In 1969, Akaike [7] used the autoregressive (AR) model for prediction. Then, the combination of the autoregressive model (AR) with the moving average (MA) model was proposed, forming the ARMA model. In 1970, Box and Pierce [8] proposed the Autoregressive Integrated Moving Average System (ARIMA) model, which remedy some drawbacks of the ARMA model for dealing with non-stationary sequences. In 1982, Engle [9] proposed the autoregressive conditional heteroskedasticity (ARCH) model to process time series volatility, in 1986 Bollerslev [10] proposed the generalized autoregressive conditional heteroscedasticity (GARCH) model. In 1987, Hull and White [11] proposed a model to solve the stochastic problem of time series volatility (SV). All these methods lay the groundwork for the development of time series forecasts.

With the development of computer technology, new prediction methods have been proposed, such as artificial neural networks [12], whose objective is to replicate the way the human brain works. Recurrent neural networks [13] are a powerful set of artificial neural network algorithms, especially useful for processing sequential data such as sound, temporal or language. Some recurrent neural networks—RNN performed better in predicting financial data and became popular, such as Long Short-Term Memory [14].

Thus, LSTM network is a specific type of RNN that has been widely applied to solve supervised learning issues [15]. It has non-linear memory cells and gate units [16], capable of processing non-stationary long-term sequences. In addition, it can extract the characteristics of financial data and reflect the characteristics of the network [17]. It offers good performance in predicting prices in the stock market, as it is an algorithm capable of identifying non-linear and hidden relationships in the data, that is, it is a supervised learning algorithm, capable of learning from a set of data in training (given a dataset, LSTM can learn a nonlinear function for regression).

Maknickiene and Maknickas [18] improved the performance of measurements in the foreign exchange market or using TM; Chen, Zhou, and Dai [19] used LSTM for Chinese design market returns and performed well. After that, several experiments with the modern LSTM network in a literary way in isolation, hybrid or combined methods with classics and financial series studies were completed, A indicates that LSTM is quite suitable for time series financial models [20].

Concretely, the LSTM network consists of three parts, including an input layer, an output layer and several hidden layers between them. The hidden layers have memory modules. The core of the memory module is the self-connecting memory cell with three ports, input, output, and forgetting. The value of each of these ports controls the flow of information in the memory module.

Information is retained by cells and memory manipulations are done by gates. Gateway: where useful information is added. Oblivion Gate: where information that is no longer useful is removed. Output Port: the task of extracting useful information from the current cell state to be displayed as a result. A vector is generated, and the information is regulated using the function that filters the values to be remembered. Vector values and regulated values are multiplied to be sent as output and input to the next cell [21].

However, designing a good network architecture for the problem studied is not a simplistic task. The model’s architecture directly interferes with its performance. The refinement process can be time consuming, as there is no formal method to perform this classification task, this is necessarily through the performance of iterative tests with several parameters, in which only the structure of greater assertiveness is maintained.

The quality of the information that the network is fed, as much or more, is reflected in the accuracy of the output layer’s response. The selection of information that will be provided to the input layer is a key factor in the design of an intelligent decision system, because even if the model is the best, it will perform poorly if the features are not well chosen. Specific methods must be used in the selection of relevant information.

3. Methodology

The aim of this study is to answer the question—using LSTM artificial neural networks, can we get reliable predictions of stock prices? For this, the 7 steps described in Figure 1, detailed below, are implemented.

Step 1—Choosing the database. Sample articles come from WoS, the world’s leading citation database.

Step 2—Using WoS Initial Search Parameters for the period from January 1, 2000 to March 31, 2022. Initially, 276 articles are identified based on variations of the keywords stock, market, LSTM, forecast, stock, predictive, regression, supervised, learn, backpropagation, supervised and backpropagation. Subsequently, exclusions are performed by applying filters in WoS itself, resulting in an intermediate sample of 127 articles, as shown in Table 1.

Figure 1. Steps of metodology.

Table 1. Evolution of the sample using WoS’ filters.

Signal	Description	Number of papers
(+)	Keywords: equal to “stock* market*” and “LSTM” or “stock* market*” and “long short-term memory” or “forecas* stock*” and “LSTM” or “forecas* stock*” and “long short-term memory” or “predictive regression*” and “LSTM” or “predictive regression*” and “long short-term memory” or “supervision *learning*” and “long short-term memory” and “backpropagation”.	276
(−)	Document type: other than “article” or “early access” or “data document”.	106
(−)	Web of Science categories: equal to “computer science artificial intelligence”, “computer science theory methods”, “computer science information systems”, “interdisciplinary applications of computer science”, “hardware architecture of computer science”, “computer science software engineering”, “economics”, “business”, “business finance”, “operations research management science”, “management”.	43
(−)	Research area: equal to “computer science”, “business economics”, “operations research management science”.	0
(−)	Language: different from “English”.	0
(=)	Intermediate sample.	127

Step 3—Exclusion of 19 of the 127 articles for not being available in the researched sources (Google Scholar, Science Direct and Web of Science) and another 09 for not being directly related to the topic of our research, namely: 01 e-commerce, 02 cryptocurrencies, 01 real estate price bubble detection, 01 hierarchical temporal memory, 01 neuromorphic vision datasets, 01 Gray Wolf-Elman optimization, 01 stock movement during the Covid-19 pandemic and 01 stock price prediction based on in morphological similarity clustering and hierarchical temporal memory. Thus, the final sample is composed of 99 articles [15] [20]-[122].

Step 4—Creation of a database and collection of articles. The 99 articles in the final sample are obtained from the following academic research databases: Web of Science, Science Direct, and Google Scholar. From its analysis, the following information is collected to capture the general data of the article: title, author name, affiliated institution and country of origin of authors/researchers, journal name, volume and issue number, homepage and page final, year of publication, country of origin of data and number of years of sample data, keywords, Digital Object Identifier (DOI), Journal of Economic Literature (JEL) and number of citations of articles in the WoS database.

Step 5—Bibliometric analysis. Through the R, RStudio, Biblioshiny and VOSviewer software, objective data from the articles are analyzed—countries, authors, keywords, institutions, etc., for the preparation and analysis of relationship/co-citation tables and maps. The analyzes carried out by both tools are complemented by the verification of the main laws of bibliometrics, namely: 1) Bradford’s Law [5]—verification of journals that produce many articles in contrast to those that produce few on a given topic, and 2) Law de Lotka [4]—identification of researchers with a higher frequency of production in a given area of knowledge.

Step 6—Reading and coding the articles. Identification of the objectives, sample, methods and contributions of the articles. In addition, they are classified and coded into categories and subcategories structured according to Table 2. Each of the 08 categories has non-exclusive subcategories. This means that the same article can be classified in more than one subcategory. Thus, the sum of the frequency count of the subcategories—for each category—is what adds up to 100%. In the coding process, as many subcategories as necessary per article are assigned.

Step 7—Systematic review. After coding the (sub)categorization matrix in Table 2—for the final sample—a frequency count of the subcategories is performed to enable the identification of knowledge gaps. Such gaps are then compared with the subcategories of category 08—paths for future studies, in order to obtain aspects that can be the object of further studies on the subject.

Table 2. Matrix of (sub) categorization.

Categories	Subcategories	Definition
1. Neural networks/ algorithms used in research	A-LSTM	Stock price projection with the—RNN LSTM.
	B-Compared to LSTM	Stock price prediction with other artificial neural networks and results compared to RNN LSTM.
	C-Combined with LSTM	Predict stock prices with blended neural networks including LSTM.
	D-Others	Other topics unrelated to subcategories 1A to 1C.
2. Types of data analyzed	A-Closing prices	Daily stock closing prices.
	B-Opening prices	Daily stock opening prices.
	C-Highest and lowest prices	Daily highest and lowest stock prices.
	D-Volumes	Stock trading volumes.
	E-Index	Daily closing of the Stock Price Index.
	F-Others	Others not related to subcategories 2A to 2E.
3. Analysis period	A-Up to 5 years	Data from 0 to 5 years.
	B-More than 5 to 10 years	Data from 5.1 to 10 years.
	C-More than 10 years	More than 10 years.
	D-Not applicable/not informed	Studies that do not inform the period of analysis.
4. Objectives	A-Tests with new neural networks models	Improved share price accuracy tested with other neural networks algorithms and/or hybrid models.
	B-Tests with other assets	Check whether using price and volatility indices of other assets (except stocks) can help predict stock prices.
	C-Sentiment Analysis	Improved accuracy in stock price projection with sentiment analysis.
	D-Others	Other topics unrelated to subcategories 7A to 7C.
5. Data origin	A-NYSE, NASDAQ, DJI, S&P, CBOE, FTSE	US Stock Exchanges.
	B-CSI, SSE, NSE, HS, SH, SZSE	China and Hong Kong Stock Exchanges.
	C-B3	Brazil Stock Exchange.
	D-TWSE	Thailand Stock Exchange.
	E-IMKB	Turkey Stock Exchange.
	F-TSE	Tehran Stock Exchange.
	G-GSE	Ghana Stock Exchange.
	H-ASX	Australia Stock Exchange.
	I-DAX	Germany Stock Exchange.
	J-KOSPI, KOSDAQ	Korea Stock Exchanges.
	K-NSE	India Stock Exchange.
	N-NIKKEI	Japan Stock Exchange.
	O-IDX	Indonesia Stock Exchange.
	P-FTSE	UK Stock Exchange.
	L-Texts	News agencies/websites, for sentiment analysis.
	M-No information or other	There is no identification of information that can be considered as inputs for the evaluation models.
6. Results	A-Outperforms compared methods	The results of the proposed model surpass the results of the compared model(s).
	B-Promising model	The results of the proposed model are promising.
	C-Others	Other results unrelated to subcategories 8A to 8B.
7. Conclusions	A-New conclusions	Presentation of new findings—adjustment to already tested neural networks models, improvement in the quality of input information, and other innovations to existing models.
	B-New perspectives	Presentation of a new theory, new models of projections, with models of isolated, hybrid or combined neural networks.
	C-Conclusions similar to works presented previously	Studies that do not present new perspectives or new conclusions.
	D-Others	Other results unrelated to subcategories 9A to 9C.
8. Pathways for future studies	A-Hybrid models with LSTM	Studies with other hybrid models using LSMT.
	B-Other ANN	Studies with other ANN, pure or hybrid.
	C-Other types of data	In addition to opening, closing, high, low and trading volume data, sentiment analysis tests with other types of news and stock data, in periods such as intraday.
	D-Data from other sources	Study the model’s performance on other Stock Exchanges.
	E-Other analysis periods	Study and test data from different periods.
	F-No path commented by the author(s)	No future path detailed by author(s).

4. Analysis of Results

Item 4.1 presents the results of the bibliometric analysis, mentioned in Step 5 of the Methodology. In turn, item 5.2 contains the results of the systematic review, whose steps are described in Steps 6 and 7 of item 3 of this study.

4.1. Bibliometric Analysis

The final sample consists of 99 articles, distributed between the years 2000 and 2021, obtained from the WoS database—see Figure 2. In this period, up to 5 articles on stock price forecasting using LSTM per year are identified.

Figure 3 shows the co-occurrence map of the most used keywords in the articles.

Again, the words model and neural networks stand out, in addition to prediction and time series.

Table 3 presents the frequency of the 151 main keywords of the study, highlighting model (18 occurrences), prediction (14 occurrences), neural networks (15 occurrences), time series (13 occurrences), index (11 occurrences), machine (10 occurrences), LSTM and neural network (with 09 occurrences each).

As for the authorship of the works, 333 authors were identified. Figure 4 shows the ranking in descending order of the 26 host countries of the institutions to which these authors are associated.

Figure 2. Annual distribution of papers.

Figure 3. Keyword co-occurrence map. Source: VOSviewer. Note: The size of the nodes represents the relevance of terms in the articles. The thickness of the lines means the strength of connection between them. Finally, the colors indicate the number of groups.

Table 3. Plus keywords.

Key words	The amount	Frequency %
Model	18	12%
Neural networks	15	10%
Prediction	14	9%
Time series	13	9%
Index	11	7%
Machine	10	7%
LSTM	9	6%
Neural network	9	6%
Volatility	7	5%
Classification	6	4%
Others	39	26%
Total	151	100%

Figure 4. Publication of articles by country to which authors are associated.

According to the RStudio software, of the 99 articles, 77 (78%) are classified as articles written by authors associated with institutions in the same country (SCP), and 22 (22%) are articles written by authors associated with institutions in different countries. countries (MCP).

Figure 5 indicates that 653 citations are related to articles written by authors associated with institutions located in China. The other citations are from authors linked to institutions of the following in Korea (380), USA (298), Pakistan (120), India (69), and the other citations, scattered among 21 other countries.

Figure 6 shows the co-citation network among journals in the final sample of 99 articles. The most cited, according to the VOSviewer software, are Expert Systems with Applications, Knowledge-Based Systems (503 co-citations), IEEE Access (235 co-citations), Neural Computing and Applications (105 co-citations) and Soft Computing (42 co-citations).

Of these, only Expert Systems with Applications stands out below, indicating that the journals that publish the most on a given topic are not necessarily the most co-cited; this fact is actually due to the relevance of each published article.

Table 4 indicates the journals in which the 99 articles of the final sample are published, through the application of Bradford’s Law [5]. The law states that there are few journals producing many articles and many journals producing few articles on a given topic. For Brookes [35], this law estimates the degree of relevance of certain academic journals that work in specific areas of knowledge. Thus, if the journals are classified in decreasing order of productivity, they can be distributed in zones with variation in the proportion 1:n:n², and so on.

Zone A is identified as the core of the disciplines, being composed of journals with 5 references or more, highlighting Expert Systems with Applications, IEEE Access, Big Data Journal and Neural Computing and Applications. Zone B presents periodicals with 2 to 4 publications, and Zone C, periodicals with a single publication.

Figure 5. Frequency of article citations in the countries of the institutions with which the authors are associated.

Figure 6. Map of co-citations between journals. Source: VOSviewer. Note: The size of the nodes represents the relevance of terms in the articles. The thickness of the lines means the strength of connection between them. The colors indicate the number of groups.

Table 4. Bradford’s law on journals.

Zone	Daily	Individual quantity	Accumulated amount	Accumulation percentage
Zone A	Expert Systems with Applications	12	12	12.1%
	IEEE Access	11	23	23.2%
	Big Data Journal	5	28	28.3%
	Neural Computing and Applications	5	33	33.3%
Zone B	Applied Smooth Computing	4	37	37.4%
	Multimedia Tools and Applications	4	41	41.4%
	Scientific Programming	4	45	45.5%
	Smooth Computing	4	49	49.5%
	Forecast Diary	3	52	52.5%
	Neurocomputing	3	55	55.6%
	PEERJ Computer Science	3	58	58.6%
	Algorithms	2	60	60.6%
	Big Data	2	62	62.6%
	Computational Economics	2	64	64.6%
	Studies and Research in Economic Computing and Economic Cybernetics	2	66	66.7%
	Quantitative finance	2	68	68.7%

Table 5 presents the ten most cited works on the RNN LSTM topic. Among them, the work by Kim and Won [20] stands out, with 235 (20.8%) of the citations and an annual average of 47.0. The article focuses on predicting the volatility of the stock price index, using a model that integrates the LSTM with several General Autoregressive models conditional Heteroskedasticity—GARCH. The second and third places are the articles by Long et al. [22] with 133 citations, an average of 33.3 per year, and Kudugunta and Ferrara [24] with 132 citations, an average of 26.4 per year.

In turn, Lotka [4] states that a small number of authors produce many works and that the production obtained by this small number of researchers is equal in quantity to the performance of the others. This law is called the inverse square law—see Equation (1).

an = a1/n², n = 1, 2, 3 (1)

In which:

an = number of authors who published n articles;

a1 = number of authors who published an article;

n = number of articles published by author.

For Equation (2), Chung and Cox [23] clarify that the number of authors with a single published article, according to Lotka’s Law, would be:

a1 = 6/π² = 0.6079 = 60.8% (2)

Table 5. The ten most cited articles.

References	Number of citations	Frequency of citations %	Total citations per year
Kim and Won [20]	235	20.8%	47.0
Long et al. [22]	133	11.7%	33.3
Kudugunta and Ferrara [24]	132	11.7%	26.4
Baek and Kim [25]	123	10.9%	24.6
Bukhari et al. [26]	120	10.6%	40.0
Pang et al. [27]	120	10.6%	40.0
Sohangir et al. [28]	93	8.2%	19.6
Jin et al. [29]	70	6.2%	23.3
Borovkova and Tsiamas [30]	57	5.0%	14.3
Xing et al. [31]	49	4.3%	9.8
Total	1,132	100.0%

Thus, an author with two published articles must have a frequency of 15.2% (0.6079/2²). For an author with three published articles it would be 6.8% (0.6079/3²) and an author with four published articles would be 6.8% (0.6079/4²).

It appears that the 99 articles in the final sample are produced by 333 authors, with one author publishing 4 articles, six authors publishing 3 articles, twenty authors publishing 2 articles and three hundred and thirty-three authors publishing a single article. 27 authors (08%), including only those who publish the most, are responsible for 62 (18.2%) publications. That said, there are not a smaller number of researchers matching the performance of the others, making it impossible to confirm Lotka’s Law.

4.2. Systematic Review

A systematic literature review seeks to identify knowledge gaps related to the topic of this study. For this, in Step 6 of Item 3—Methodology, a (sub)categorization matrix is defined—see Table 2. Categories and subcategories are identified for each of the 99 articles in the final sample. In this way, the frequency count is made in relation to the total of the subcategories and not to the total of the 99 articles.

The subcategories that have the potential to be prioritized in future research are highlighted. In category 1, Neural Networks/Algorithms Used in the Research, the theme “Forecasting stock prices with other artificial neural networks and results compared to RNN LSTM” is the most relevant (65%), followed by “Forecasting stock prices with blended neural networks, including LSTM” (22%) and Share Price Projection with RNN LSTM (23%).

Regarding category 2 “Type of Data Analyzed”, the closing price was the most used data in the surveys, alone (22%) or associated with other data (61%).

Regarding the period of analysis, category 3, 37% of the articles were concentrated in periods of up to 05 years, 37% in periods of 05 to 10 years, 16% in periods of more than 10 years, and 09% did not inform.

In category 4, the objectives of the papers are highlighted. 74% of them project the price of shares with LSTM alone or associated with the most varied RNN (without relevant concentrations); 19% test hybrid neural networks and sentiment analysis models.

Category 5 indicates that 21% of the papers are exclusively based on data from US stock exchanges, 20% exclusively use data from Chinese stock exchanges, 08% jointly use data from American and Chinese stock exchanges and other associations. 38% of articles do not use data from the US or Chinese stock exchanges, but from the stock exchanges of Brazil, Thailand, Turkey, Tehran, Ghana, Australia, Germany, Korea, India, Japan, Indonesia or the United Kingdom, in association or in isolation. Thus, it can be seen that there is plenty of room for studies based on Stock Exchanges in other developed and/or developing countries. 08% perform sentiment analysis, that is, they use texts exposed in the media, and not data from Stock Exchanges.

In turn, category 6 presents the results of the studies carried out. In 57% of them, the proposed models outperform the results of the models to which they were compared; in 42% the results are defined as promising, indicating that they can be improved.

According to category 7, 57% of the studies present adjustments in already tested neural network models, with improvement in the quality of input information and/or other innovations in existing models. 42% of the articles present new theories or new models of projections, using simple, hybrid or combined neural networks.

Finally, category 8 indicates paths for future studies, that is, knowledge gaps according to the authors of the 99 papers in the final sample. In 27% of the papers, the authors suggest the use of innovative data, such as other news sources and/or stock data, in different time periods, such as intraday. 22% suggest studies with hybrid models of LSTM not yet tested, associated or not with other neural networks, other types of data, other sources and other periods. 10% suggest studies with other types of neural networks, pure or hybrid.

5. Conclusion

Publications on this topic are concentrated from 2020 onwards. The keywords most associated with these studies are model, neural networks, prediction and time series. 333 authors wrote on the subject between 2018 and March 2022; 43 of the 99 articles published in this period are associated with Chinese institutions. The journals that publish the most significant articles on the topic are Expert Systems with Applications, IEEE Access, Big Data Journal and Neural Computing and Applications. The most cited article is by Kim and Won [20], Stock Price Index Volatility Prediction: A Hybrid Model Integrating LSTM with Various GARCH-Type Models, which studies the volatility of Kospi 200 stock index returns and capitalization of the stock market in South Korea, cited 235 times. The daily closing price of shares is the most analyzed type of data, and studies are still concentrated on American (21%) and Chinese (20%) stock exchanges. 57% of the studies present adjustments to already tested neural network models and 42% present new theories or new projection models. In 27% of the articles, the authors suggest future studies with news sources, other stock data, or the use of different time series. 22% suggest studies with hybrid models of LSTM not yet tested, associated or not with other neural networks, other data, other sources and periods.

Conflicts of Interest

The authors declare no conflicts of interest regarding the publication of this paper.

References

[1]	Lunga, D. and Marwala, T. (2006) Online Forecasting of Stock Market Movement Direction Using the Improved Incremental Algorithm. In: King, I., Wang, J., Chan, L.W. and Wang, D., Eds., Neural Information Processing, Vol. 4234, Springer, Berlin, 440-449. https://doi.org/10.1007/11893295_49
[2]	Belciug, S. and Sandita, A. (2017) Business Intelligence: Statistics in Predicting Stock Market. Annals of the University of Craiova, Mathematics and Computer Science Series, 44, 292-298.
[3]	Demirel, U., Cam, H. and Unlu, R. (2021) Predicting Stock Prices Using Machine Learning Methods and Deep Learning Algorithms: The Sample of the Istanbul Stock Exchange. Gazi University Journal of Science, 34, 63-82. https://doi.org/10.35378/gujs.679103
[4]	Lotka, A.J. (1926) The Frequency Distribution of Scientific Productivity. Journal of the Washington Academy of Sciences, 16, 317-323.
[5]	Bradford, S.C. (1934) Sources of Information on Scientific Subjects. Engineering: An Illustrated Weekly Journal, 137, 85-86.
[6]	Frasconi, P., Gori, M., Maggini, M. and Soda, G. (1995) Unified Integration of Explicit Knowledge and Learning by Example in Recurrent Networks. IEEE Transactions on Knowledge and Data Engineering, 7, 340-346. https://doi.org/10.1109/69.382304
[7]	Akaike, H. (1969) A Method of Statistical Identification os Discrete Time Parameter Linear Systems. The Institute of Statistical Mathematics, 21, 225-242. https://doi.org/10.1007/BF02532250
[8]	Box, G.E. and Pierce, D.A. (1970) Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models. Journal of the American Statistical Association, 65, 1509-1526. https://doi.org/10.1080/01621459.1970.10481180
[9]	Engle, R.F. (1982) Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation. Econometrica, 50, 987-1007. https://doi.org/10.2307/1912773
[10]	Bollersley, T. (1986) Generalized Autoregressive Conditional Heteroskedasticity. Journal of Econometrics, 31, 307-327. https://doi.org/10.1016/0304-4076(86)90063-1
[11]	Hull, J. and White, A. (1987) The Pricing of Options on Assets with Stochastic Volatilities. Journal of Finance, 42, 281-300. https://doi.org/10.1111/j.1540-6261.1987.tb02568.x
[12]	Paluch, M. and Jackowska-Strumillo, L. (2018) Hybrid Models Combining Technical and Fractal Analysis with ANN for Short-Term Prediction of Close Values on the Warsaw Stock Exchange. Applied Sciences, 8, Article 2473. https://doi.org/10.3390/app8122473
[13]	Gao, T. and Chai, Y. (2018) Improving Stock Closing Price Prediction Using Recurrent Neural Network and Technical Indicators. Neural Computation, 30, 2833- 2854. https://doi.org/10.1162/neco_a_01124
[14]	Hochreiter, S. and Schmidhuber, J. (1997) Long Short-Term Memory. Neural Computation, 9, 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735
[15]	Li, Z.X. and Tam, V. (2018) A Machine Learning View on Momentum and Reversal Trading. Algorithms, 11, Article 170. https://doi.org/10.3390/a11110170
[16]	Greff, K., Srivastava, R.K., Koutnik, J., Steunebrink, B.R. and Schmidhuber, J. (2016) LSTM: A Search Space Odyssey. IEEE Transactions on Neural Networks and Learning Systems, 28, 2222-2232. https://doi.org/10.1109/TNNLS.2016.2582924
[17]	Zeng, Y. and Liu, X. (2018) A-Stock Price Fluctuation Forecast Model Based on LSTM. 14th International Conference on Semantics, Knowledge and Grids (SKG), Guangzhou, 12-14 September 2018, 261-264. https://doi.org/10.1109/SKG.2018.00044
[18]	Maknickiene, N. and Maknickas, A. (2012) Application of Neural Network for Forecasting of Exchange Rates and Forex Trading. The 7th International Scientific Conference “Business and Management”, Vilnius, 10-11 May 2012, 122-127. https://doi.org/10.3846/bm.2012.017
[19]	Chen, K., Zhou Y. and Fangyan, D. (2015) A LSTM-Based Method for Stock Returns Prediction: A Case Study of China Stock Market. IEEE International Conference on Big Data, Santa Clara, 29 October 2015-01 November 2015, 2823-2824. https://doi.org/10.1109/BigData.2015.7364089
[20]	Kim, H.Y. and Won, C.H. (2018) Forecasting the Volatility of Stock Price Index: A Hybrid Model Integrating LSTM with Multiple GARCH-Type Models. Expert Systems with Applications, 103, 25-37. https://doi.org/10.1016/j.eswa.2018.03.002
[21]	Liu, X., Liu, H., Guo, Q. and Zhang, C.M. (2020) Adaptive Wavelet Transform Model for Time Series Data Prediction. Soft Computing, 24, 5877-5884. https://doi.org/10.1007/s00500-019-04400-w
[22]	Long, W., Lu, Z.C. and Cui, L.X. (2019) Deep Learning-Based Feature Engineering for Stock Price Movement Prediction. Knowledge-Based Systems, 164, 163-173. https://doi.org/10.1016/j.knosys.2018.10.034
[23]	Chung, K.H. and Cox, R.A.K. (1990) Productivity Patterns in the Finance Literature: A Study of Bibliometric Distributions. Journal of Finance, 45, 301-309. https://doi.org/10.1111/j.1540-6261.1990.tb05095.x
[24]	Kudugunta, S. and Ferrara, E. (2018) Deep Neural Networks for Bot Detection. Information Sciences, 467, 312-322. https://doi.org/10.1016/j.ins.2018.08.019
[25]	Baek, Y. and Kim, H.Y. (2018) ModAugNet: A New Forecasting Framework for Stock Market Index Value with an Overfitting Prevention LSTM Module and a Prediction LSTM Module. Expert Systems with Applications, 113, 457-460. https://doi.org/10.1016/j.eswa.2018.07.019
[26]	Bukhari, A.H., Raja, M.A.Z., Sulaiman, M., Islam, S., Shoaib, M. and Kumam, P. (2020) Fractional Neuro-Sequential ARFIMA-LSTM for Financial Market Forecasting. IEEE Access, 8, 71326-71338. https://doi.org/10.1109/ACCESS.2020.2985763
[27]	Pang, X.W., Zhou, Y.Q., Wang, P., Lin, W.W. and Chang, V. (2020) An Innovative Neural Network Approach for Stock Market Prediction. Journal of Supercomputing, 76, 2098-2118. https://doi.org/10.1007/s11227-017-2228-y
[28]	Sohangir, S., Wang, D.D., Pomeranets, A. and Khoshgoftaar, T.M. (2018) Big Data: Deep Learning for Financial Sentiment Analysis. Journal of Big Data, 5, Article No. 3. https://doi.org/10.1186/s40537-017-0111-6
[29]	Jin, Z.G., Yang, Y. and Liu, Y.H. (2020) Stock Closing Price Prediction Based on Sentiment Analysis and LSTM. Neural Computing & Applications, 32, 9713-9729. https://doi.org/10.1007/s00521-019-04504-2
[30]	Borovkova, S. and Tsiamas, I. (2019) An Ensemble of LSTM Neural Networks for High-Frequency Stock Market Classification. Journal of Forecasting, 1-27. https://doi.org/10.2139/ssrn.3202313
[31]	Xing, F.Z., Cambria, E. and Welsch, R.E. (2018) Intelligent Asset Allocation via Market Sentiment Views. IEEE Computational Intelligence Magazine, 13, 25-34. https://doi.org/10.1109/MCI.2018.2866727
[32]	AbdelKawy, R., Abdelmoez, W.M. and Shoukry, A. (2021) The Synchronous Deep Reinforcement Learning Model for Automated Multi-Stock Trading. Progress in Artificial Intelligence, 10, 83-97. https://doi.org/10.1007/s13748-020-00225-z
[33]	Affonso, F., Dias, T.M.R. and Pinto, A.L. (2021) Financial Times Series Forecasting of Clustered Stocks. Mobile Networks & Applications, 26, 256-265. https://doi.org/10.1007/s11036-020-01647-8
[34]	Althelaya, K.A., Mohammed, S.A. and El-Alfy, E.M. (2021) Combining Deep Learning and Multiresolution Analysis for Stock Market Forecasting. IEEE Access, 9, 13099-13111. https://doi.org/10.1109/ACCESS.2021.3051872
[35]	Brookes, B.C. (1969) Bradford’s Law and the Bibliography of Science. Nature, 222, 953-956. https://doi.org/10.1038/224953a0
[36]	Bucci, A. (2020) Realized Volatility Forecasting with Neural Networks. Journal of Financial Econometrics, 18, 502-531. https://doi.org/10.1093/jjfinec/nbaa008
[37]	Bucci, A. (2020) Cholesky-ANN Models for Predicting Multivariate Realized Volatility. Journal of Forecasting, 39, 865-876. https://doi.org/10.1002/for.2664
[38]	Budiharto, W. (2021) Data Science Approach to Stock Prices Forecasting in Indonesia during COVID-19 Using Long Short-Term Memory (LSTM). Journal of Big Data, 8, Article No. 47. https://doi.org/10.1186/s40537-021-00430-0
[39]	De Oliveira Carosia, A.E., Coelho, G.P. and Da Silva, A.E.A. (2021) Investment Strategies Applied to the Brazilian Stock Market: A Methodology Based on Sentiment Analysis with Deep Learning. Expert Systems with Applications, 184, Article ID: 115470. https://doi.org/10.1016/j.eswa.2021.115470
[40]	Chacon, H.D., Kesici, E. and Najafirad, P. (2020) Improving Financial Time Series Prediction Accuracy Using Ensemble Empirical Mode Decomposition and Recurrent Neural Networks. IEEE Access, 8, 117133-117145. https://doi.org/10.1109/ACCESS.2020.2996981
[41]	Chang, V.C., Man, X.W., Xu, Q.W. and Hsu, C.H. (2021) Pairs Trading on Different Portfolios Based on Machine Learning. Expert Systems, 38, e12649. https://doi.org/10.1111/exsy.12649
[42]	Chen, Q., Zhang, W.Y. and Lou, Y. (2020) Forecasting Stock Prices Using a Hybrid Deep Learning Model Integrating Attention Mechanism, Multi-Layer Perceptron, and Bidirectional Long-Short Term Memory Neural Network. IEEE Access, 8, 117365-117376. https://doi.org/10.1109/ACCESS.2020.3004284
[43]	Chen, S. and Ge, L. (2019) Exploring the Attention Mechanism in LSTM-Based Hong Kong Stock Price Movement Prediction. Quantitative Finance, 19, 1507-1515. https://doi.org/10.1080/14697688.2019.1622287
[44]	Chen, Y., Fang, R.X., Liang, T., Sha, Z.Y., Li, S.C., Yi, Y.G., Zhou, W. and Song, H.L. (2021) Stock Price Forecast Based on CNN-BiLSTM-ECA Model. Scientific Programming, 2021, Article ID: 2446543. https://doi.org/10.1155/2021/2446543
[45]	Chen, Y.C. and Huang, W.C. (2021) Building a Stock-Price Forecast CNN Model with Gold and Crude Oil Indicators. Applied Soft Computing, 112, Article ID: 107760. https://doi.org/10.1016/j.asoc.2021.107760
[46]	Dey, P., Hossain, E., Hossain, M.I., Chowdhury, M.A., Alam, M.S., Hossain, M.S. and Andersson, K. (2021) Comparative Analysis of Recurrent Neural Networks in Stock Price Prediction for Different Frequency Domains. Algorithms, 14, Article 251. https://doi.org/10.3390/a14080251
[47]	Ding, G.Y. and Qin, L.X. (2020) Study on the Prediction of Stock Price Based on the Associated Network Model of LSTM. International Journal of Machine Learning and Cybernetics, 11, 1307-1317. https://doi.org/10.1007/s13042-019-01041-1
[48]	Dong, S.T., Wang, J.X., Luo, H.Z., Wang, H.D. and Wu, F.X. (2021) The Dynamic Predictor Selection Algorithm for Predicting Stock Market Movement. Expert Systems with Applications, 186, Article ID: 115836. https://doi.org/10.1016/j.eswa.2021.115836
[49]	Gao, Y., Wang, R. and Zhou, E.M. (2021) Stock Prediction Based on Optimized LSTM and GRU Models. Scientific Programming, 2021, Article ID: 4055281. https://doi.org/10.1155/2021/4055281
[50]	Garcia-Vega, S., Zeng, X.J. and Keane, J. (2020) Stock Returns Prediction Using Adaptive Kernel Filtering within a Stock Market Interdependence Approach. Expert Systems with Applications, 160, Article ID: 113668. https://doi.org/10.1016/j.eswa.2020.113668
[51]	Gite, S., Khatavkar, H., Kotecha, K., Srivastava, S., Maheshwari, P. and Pandey, N. (2021) Explainable Stock Prices Prediction from Financial News Articles Using Sentiment Analysis. PEERJ Computer Science, 7, e340. https://doi.org/10.7717/peerj-cs.340
[52]	Gu, W.T., Zhang, L.H., Xi, H.J., Zheng, S.H. (2021) Stock Prediction Based on News Text Analysis. Journal of Advanced Computational Intelligence and Intelligent Informatics, 25, 581-591. https://doi.org/10.20965/jaciii.2021.p0581
[53]	Gunduz, H. (2021) An Efficient Stock Market Prediction Model Using Hybrid Feature Reduction Method Based on Variational Autoencoders and Recursive Feature Elimination. Financial Innovation, 7, Article No. 28. https://doi.org/10.1186/s40854-021-00243-3
[54]	Han, J.J. and Kim, H.J. (2021) Prediction of Investor-Specific Trading Trends in South Korean Stock Markets Using BiLSTM Prediction Model Based on Sentiment Analysis of Financial News Articles. Journal of Behavioral Finance. https://doi.org/10.1080/15427560.2021.1995735
[55]	Hansun, S. and Young, J.C. (2021) Predicting LQ45 Financial Sector Indices Using RNN-LSTM. Journal of Big Data. https://doi.org/10.21203/rs.3.rs-471941/v1
[56]	He, B.T. and Kita, E. (2021) The Application of Sequential Generative Adversarial Networks for Stock Price Prediction. Review of Socionetwork Strategies, 15, 455-470. https://doi.org/10.1007/s12626-021-00097-2
[57]	Huang, X. and Song, H.L. (2021) Investor Sentiment Combined with Multisource Information to Predict Stock Prices: An Analysis of China’s A-Share Market. Scientific Programming, 2021, Article ID: 9094032. https://doi.org/10.1155/2021/9094032
[58]	Huang, Y.S., Gao, Y.L., Gan, Y. and Ye, M. (2021) A New Financial Data Forecasting Model Using Genetics Algorithm and Long Short Term Memory Network. Neurocomputing, 425, 207-218. https://doi.org/10.1016/j.neucom.2020.04.086
[59]	Ji, Y., Liew, A.W.C and Yang, L.X. (2021) The Novel Improved Particle Swarm Optimization with Long-Short Term Memory Hybrid Model for Stock Indices Forecast. IEEE Access, 9, 23660-23671. https://doi.org/10.1109/ACCESS.2021.3056713
[60]	Jing, N., Wu, Z. and Wang, H.F. (2021) A Hybrid Model Integrating Deep Learning with Investor Feeling Analysis for Stock Price Prediction. Expert Systems with Applications, 178, Article ID: 115019. https://doi.org/10.1016/j.eswa.2021.115019
[61]	Kelotra, A. and Pandey, P. (2020) Stock Market Prediction Using Optimized Deep-ConvLSTM Model. Big Data, 8, 5-24. https://doi.org/10.1089/big.2018.0143
[62]	Khalil, F. and Pipa, G. (2021) Is Deep-Learning and Natural Language Processing Transcending the Financial Forecasting? Investigation through Lens of News Analytic Process. Computational Economics, 60, 147-171. https://doi.org/10.1007/s10614-021-10145-2
[63]	Kilimci, Z.H. and Duvar, R. (2020) An Efficient Word Embedding and Deep Learning Based Model to Forecast the Direction of Stock Exchange Market Using Twitter and Financial News Sites: A Case of Istanbul Stock Exchange (BIST 100). IEEE Access, 8, 188186-188198. https://doi.org/10.1109/ACCESS.2020.3029860
[64]	Kim, S., Ku, S., Chang, W. and Song, J.W. (2020) Predicting the Direction of US Stock Prices Using Effective Transfer Entropy and Machine Learning Techniques. IEEE Access, 8, 111660-111682. https://doi.org/10.1109/ACCESS.2020.3002174
[65]	Ko, C.R. and Chang, H.T. (2021) LSTM-Based Sentiment Analysis for Stock Price Forecast. PEERJ Computer Science, 7, e408. https://doi.org/10.7717/peerj-cs.408
[66]	Kumar, A., Alsadoon, A., Prasad, P.W.C., Abdullah, S., Rashid, T.A., Pham, D.T.H. and Nguyen, T.Q.V. (2022) Generative Adversarial Network (GAN) and Enhanced Root Mean Square Error (ERMSE): Deep Learning for Stock Price Movement Prediction. Multimedia Tools and Applications, 81, 3995-4013. https://doi.org/10.1007/s11042-021-11670-w
[67]	Kumar, K. and Haider, M.T.U. (2021) Enhanced Prediction of Intra-Day Stock Market Using Metaheuristic Optimization on RNN-LSTM Network. New Generation Computing, 39, 231-272. https://doi.org/10.1007/s00354-020-00104-0
[68]	Kumar, R., Kumar, P. and Kumar, Y. (2021) Integrating Big Data Driven Sentiments Polarity and ABC-Optimized LSTM for Time Series Forecasting. Multimedia Tools and Applications, 81, 34595-34614. https://doi.org/10.1007/s11042-021-11029-1
[69]	Lee, M.C., Chang, J.W., Hung, J.S.C. and Chen, B.L. (2021) Exploring the Effectiveness of Deep Neural Networks with Technical Analysis Applied to Stock Market Prediction. Computer Science and Information Systems, 18, 401-418. https://doi.org/10.2298/CSIS200301002L
[70]	Li, M.G., Li, W.R., Wang, F., Jia, X.J. and Rui, G.W. (2021) Applying BERT to Analyze Investor Sentiment in Stock Market. Neural Computing and Applications, 33, 4663-4676. https://doi.org/10.1007/s00521-020-05411-7
[71]	Li, Q., Tan, J.H., Wang, J. and Chen, H. (2021) A Multimodal Event-Driven LSTM Model for Stock Prediction Using Online News. IEEE Transactions on Knowledge and Data Engineering, 33, 3323-3337. https://doi.org/10.1109/TKDE.2020.2968894
[72]	Li, Y.L., Bu, H., Li, J.H. and Wu, J.J. (2020) The Role of Text-Extracted Investor Sentiment in Chinese Stock Price Prediction with the Enhancement of Deep Learning. International Journal of Forecasting, 36, 1541-1562. https://doi.org/10.1016/j.ijforecast.2020.05.001
[73]	Lin, Y., Yan, Y., Xu, J.L., Liao, Y. and Ma, F. (2021) Forecasting Stock Index Price Using the CEEMDAN-LSTM Model. North American Journal of Economics and Finance, 57, Article ID: 101421. https://doi.org/10.1016/j.najef.2021.101421
[74]	Liu, Y., Yang, C., Huang, K., Gui, W. (2020) Non-Ferrous Metals Price Forecasting Based on Variational Mode Decomposition and LSTM Network. Knowledge-Based Systems, 188, Article ID: 105006. https://doi.org/10.1016/j.knosys.2019.105006
[75]	Liu, Y.Z., Yu, X.L., Wu, Y.H. and Song, S.H. (2021) Forecasting Variation Trends of Stocks via Multiscale Feature Fusion and Long Short-Term Memory Learning. Scientific Programming, 2021, Article ID: 5113151. https://doi.org/10.1155/2021/5113151
[76]	Ma, C., Zhang, J.S., Liu, J.M., Ji, L.Z. and Gao, F. (2021) A Parallel Multi-Module Deep Reinforcement Learning Algorithm for Stock Trading. Neurocomputing, 449, 290-302. https://doi.org/10.1016/j.neucom.2021.04.005
[77]	Ma, Y.L., Han, R.Z. and Wang, W.Z. (2020) Prediction-Based Portfolio Optimization Models Using Deep Neural Networks. IEEE Access, 8, 115393-115405. https://doi.org/10.1109/ACCESS.2020.3003819
[78]	Ma, Y.L., Han, R.Z. and Wang, W.Z. (2021) Portfolio Optimization with Return Prediction Using Deep Learning and Machine Learning. Expert Systems with Applications, 165, Article ID: 113973. https://doi.org/10.1016/j.eswa.2020.113973
[79]	Makinen, Y., Kanniainen, J., Gabbouj, M. and Iosifidis, A. (2019) Forecasting Jump Arrivals in Stock Prices: New Attention-Based Network Architecture Using Limit Order Book Data. Quantitative Finance, 19, 2033-2050, https://doi.org/10.1080/14697688.2019.1634277
[80]	McNelis, P.D. (2005) Neural Networks in Finance. Academic Press, Cambridge, MA. https://doi.org/10.1016/B978-012485967-8.50002-6
[81]	Mehta, P., Pandya, S. and Kotecha, K. (2021) Harvesting Social Media Sentiment Analysis to Enhance Stock Market Prediction Using Deep Learning. PeerJ Computer Science, 7, e476. https://doi.org/10.7717/peerj-cs.476
[82]	Nabipour, M., Nayyeri, P., Jabani, H., Shahab, S. and Mosavi, A. (2020) Predicting Stock Market Trends Using Machine Learning and Deep Learning Algorithms Via Continuous and Binary Data; A Comparative Analysis. IEEE Access, 8, 150199- 150212. https://doi.org/10.1109/ACCESS.2020.3015966
[83]	Nguyen, H.T., Tran, T.B. and Bui, P.H.D. (2021) Concurrency and Computation-Practice & Experience, e6437. https://doi.org/10.1002/cpe.6437
[84]	Nobre, R.A., Do Nascimento, K.C., Vargas, P.A., Valejo, A.D.B., Pessin, G., Villas, L.A. and Rocha, G.P. (2022) Aurora: An Autonomous Agent-Oriented Hybrid Trading Service. Neural Computing & Applications, 34, 2217-2232. https://doi.org/10.1007/s00521-021-06508-3
[85]	Nti, I.K., Adekoya, A.F. and Weyori, B.A. (2021) A Novel Multi-Source Information-Fusion Predictive Framework Based on Deep Neural Networks for Accuracy Enhancement in Stock Market Prediction. Journal of Big Data, 8, Article No. 17. https://doi.org/10.1186/s40537-020-00400-y
[86]	Pardoe, I. (2006) Applied Regression Modeling: A Business Approach. John Wiley & Sons Inc., Hoboken. https://doi.org/10.1002/9781118274415
[87]	Park, D. and Ryu, D. (2021) A Machine Learning-Based Early Warning System for the Housing and Stock Markets. IEEE Access, 9, 85566-85572. https://doi.org/10.1109/ACCESS.2021.3077962
[88]	Park, H.J., Kim, Y. and Kim, H.Y. (2022) Stock Market Forecasting Using a Multi-Task Approach Integrating Long Short-Term Memory and the Random Forest Framework. Applied Soft Computing, 114, Article ID: 108106. https://doi.org/10.1016/j.asoc.2021.108106
[89]	Prachyachuwong, K. and Vateekul, P. (2021) Stock Trend Prediction Using Deep Learning Approach on Technical Indicator and Industrial Specific Information. Information, 12, Article 250. https://doi.org/10.3390/info12060250
[90]	Qiu, Y., Yang, H.Y., Lu, S. and Chen, W. (2020) A Novel Hybrid Model Based on Recurrent Neural Networks for Stock Market Timing. Soft Computing, 24, 15273- 15290. https://doi.org/10.1007/s00500-020-04862-3
[91]	Ray, P., Ganguli, B. and Chakrabarti, A. (2021) A Hybrid Approach of Bayesian Structural Time Series with LSTM to Identify the Influence of News Sentiment on Short-Term Forecasting of Stock Price. IEEE Transactions on Computational Social Systems, 8, 1153-1162. https://doi.org/10.1109/TCSS.2021.3073964
[92]	Rezaei, H., Faaljou, H. and Mansourfar, G. (2021) Intelligent Asset Allocation Using Predictions of Deep Frequency Decomposition. Expert Systems with Applications, 186, Article ID: 115715. https://doi.org/10.1016/j.eswa.2021.115715
[93]	Serrano, W. (2022) The Random Neural Network in Price Predictions. Neural Computing & Applications, 34, 855-873. https://doi.org/10.1007/s00521-021-05903-0
[94]	Sharaf, M., Hemdan, E.E., El-Sayed, A. and El-Bahnasawy, N.A. (2021) StockPred: A Framework for Stock Price Prediction. Multimedia Tools and Applications, 80, 17923-17954. https://doi.org/10.1007/s11042-021-10579-8
[95]	Sharma, S., Elvira, V., Chouzenoux, E. and Majumdar, A. (2021) Recurrent Application in Stock Forecasting. Neurocomputing, 450, 1-13. https://doi.org/10.1016/j.neucom.2021.03.111
[96]	Shu, W.W. and Gao, Q. (2020) Forecasting Stock Price Based on Frequency Components by EMD and Neural Networks. IEEE Access, 8, 206388-206395. https://doi.org/10.1109/ACCESS.2020.3037681
[97]	Song, D., Busogi, M., Baek, A.M.C. and Kim, N. (2020) Forecasting Stock Market Index Based on Pattern-Driven Long Short-Term Memory. Economic Computation and Economic Cybernetics Studies and Research, 54, 25-41.
[98]	Song, T. and Yan, X.S. (2021) Dynamic Adjustment of Stock Position Based on Hybrid Deep Neural Network. Journal of Ambient Intelligence and Humanized Computing, 12, 10073-10089. https://doi.org/10.1007/s12652-020-02768-4
[99]	Su, Z., Xie, H.L. and Han, L. (2021) Multi-Factor RFG-LSTM Algorithm for Stock Sequence Predicting. Computational Economics, 57, 1041-1058. https://doi.org/10.1007/s10614-020-10008-2
[100]	Thakkar, A. and Chaudhari, K. (2020) Predicting Stock Trend Using an Integrated Term Frequency-Inverse Document Frequency-Based Feature Weight Matrix with Neural Networks. Applied Soft Computing Journal, 96, Article ID: 106684. https://doi.org/10.1016/j.asoc.2020.106684
[101]	Touzani, Y. and Douzi, K. (2021) An LSTM and GRU Based Trading Strategy Adapted to the Moroccan Market. Journal of Big Data, 8, Article No. 126. https://doi.org/10.1186/s40537-021-00512-z
[102]	Troiano, L., Villa, E.M. and Loia, V. (2018) Replicating a Trading Strategy by Means of LSTM for Financial Industry Applications. IEEE Transactions on Industrial Informatics, 14, 3226-3234. https://doi.org/10.1109/TII.2018.2811377
[103]	Wang, J.J., Feng, C.C., He, J.J., Feng, L. and Li, Y. (2020) A Novel Multi-Factor Stock Index Prediction Approach Using Principal Component Analysis, Feature Classification and Two-Stage Long Short-Term Memory Network with Residual Correction. Economic Computation and Economic Cybernetics Studies and Research, 54, 95-100.
[104]	Wang, J.J., He, J.J., Feng, C.C., Feng, L. and Li, Y. (2021) Stock Index Prediction and Uncertainty Analysis Using Multi-Scale Nonlinear Ensemble Paradigm of Optimal Feature Extraction, Two-Stage Deep Learning and Gaussian Process Regression. Applied Soft Computing, 113, Article ID: 107898. https://doi.org/10.1016/j.asoc.2021.107898
[105]	Wang, W.J., Tang, Y., Xiong, J. and Zhang, Y.C. (2021) Stock Market Index Prediction Based on Reservoir Computing Models. Expert Systems with Applications, 178, Article ID: 115022. https://doi.org/10.1016/j.eswa.2021.115022
[106]	Wu, D.M., Wang, X.L. and Wu, S.C. (2022) Jointly Modeling Transfer Learning of Industrial Chain Information and Deep Learning for Stock Prediction. Expert Systems with Applications, 191, Article ID: 116257. https://doi.org/10.1016/j.eswa.2021.116257
[107]	Wu, J.M.T., Li, Z.C., Herencsar, N., Vo, B. and Lin, J.C.W. (2021) A Graph-Based CNN-LSTM Stock Price Prediction Algorithm with Leading Indicators. Multimedia Systems. https://doi.org/10.1007/s00530-021-00758-w
[108]	Wu, J.M.T., Sun, L.Y., Srivastava, G. and Lin, J.C.W. (2021) A Long Short-Term Memory Network Stock Price Prediction with Leading Indicators. Big Data, 9, 343-357. https://doi.org/10.1089/big.2020.0391
[109]	Wu, J.M.T., Sun, L.Y., Srivastava, G. and Lin, J.C.W. (2021) A Novel Synergetic LSTM-GA Stock Trading Suggestion System in Internet of Things. Mobile Information Systems, 2021, Article ID: 6706345. https://doi.org/10.1155/2021/6706345
[110]	Yan, X., Wang, W.H. and Chang, M. (2021) Research on Financial Assets Transaction Prediction Model Based on LSTM Neural Network. Neural Computing & Applications, 33, 257-270. https://doi.org/10.1007/s00521-020-04992-7
[111]	Yang, S.G. (2021) A Novel Study on Deep Learning Framework to Predict and Analyze the Financial Time Series Information. Future Generation Computer Systems, 125, 812-819. https://doi.org/10.1016/j.future.2021.07.017
[112]	Yang, Y.J., Yang, Y.M. and Zhou, W. (2021) Research on a Hybrid Prediction Model for Stock Price Based on Long Short-Term Memory and Variational Mode Decomposition. Soft Computing, 25, 13513-13531. https://doi.org/10.1007/s00500-021-06122-4
[113]	Yap, K.L., Lau, W.Y. and Ismail, I. (2021) Deep Learning Neural Network for the Prediction of Asian Tiger Stock Markets. International Journal of Financial Engineering, 8, Article ID: 2150040. https://doi.org/10.1142/S2424786321500407
[114]	Yeung, J.F.K.A., Wei, Z.-K., Chan, K.Y., Lau, H.Y.K. and Yiu, K.-F.C. (2020) Jump Detection in Financial Time Series Using Machine Learning Algorithms. Soft Computing, 24, 1789-1801. https://doi.org/10.1007/s00500-019-04006-2
[115]	Yildiz, Z.C. and Yildiz, S.B. (2020) The Portfolio Construction Framework Using LSTM-Based Stock Markets Forecasting. International Journal of Finance & Economics, 27, 2356-2366.
[116]	Zhang, D.X. and Cai, E.G. (2021) Improving Stock Price Forecasting Using a Large Volume of News Headline Text. Computers, Materials & Continua, 69, 3931-3943. https://doi.org/10.32604/cmc.2021.012302
[117]	Zhang, W.G., Gong, X., Wang, C. and Ye, X. (2021) Predicting Stock Market Volatility Based on Textual Sentiment: A Nonlinear Analysis. Journal of Forecasting, 40, 1479-1500. https://doi.org/10.1002/for.2777
[118]	Zhang, Y.A., Yan, B.B. and Aasma, M. (2020) A Novel Deep Learning Framework: Prediction and Analysis of Financial Time Series Using CEEMD and LSTM. Expert Systems with Applications, 159, Article ID: 113609. https://doi.org/10.1016/j.eswa.2020.113609
[119]	Zhao, Y., Shen, Y.Y. and Huang, Y. (2019) DMDP: A Dynamic Multi-Source Default Probability Prediction Framework. Data Science and Engineering, 4, 3-13. https://doi.org/10.1007/s41019-019-0085-9
[120]	Zhao, Y.Z. (2021) A Novel Stock Index Intelligent Prediction Algorithm Based on Attention-Guided Deep Neural Network. Wireless Communications and Mobile Computing, 2021, Article ID: 6210627. https://doi.org/10.1155/2021/6210627
[121]	Zolfaghari, M. and Gholami, S. (2021) A Hybrid Approach of Adaptive Wavelet Transform Long Short-Term Memory and ARIMA-GARCH Family Models for the Stock Index Prediction. Expert Systems with Applications, 182, Article ID: 115149. https://doi.org/10.1016/j.eswa.2021.115149
[122]	Dami, S. and Esterabi, M. (2021) Predicting Stock Returns of Tehran Exchange Using LSTM Neural Network and Feature Engineering Technique. Multimedia Tools and Applications, 80, 19947-19970. https://doi.org/10.1007/s11042-021-10778-3