Stock market trading is an activity in which investors need fast and accurate information to make effective decisions.But the fact is that forecasting stock prices by using various models has been suffering from low a...Stock market trading is an activity in which investors need fast and accurate information to make effective decisions.But the fact is that forecasting stock prices by using various models has been suffering from low accuracy,slow convergence,and complex parameters.This study aims to employ a mixed model to improve the accuracy of stock price prediction.We present how to use a random walk based on jump-diffusion,to obtain stock predictions with a good-fitting degree by adjusting different parameters.Aimed at getting better parameters and then using the time series model to predict the data,we employed the time series model to smooth the sequence utilizing logarithm and difference,which successfully resulted in drawing the auto-correlation figure and partial the auto-correlation figure.According to the comparative analysis,which focuses on checking the mean absolute error,including root mean square error and R square evaluation index,we have drawn a clear conclusion that our mixed model prediction effect is relatively good.In the context of Chinese stocks,the hybrid random walk model is very suitable for predicting stocks.It can“interpret”the randomness of stocks very well,and it also has an unparalleled prediction effect compared with other models.Based on the time series model’s application in this paper,the abovementioned series is more suitable for predicting trends.展开更多
Previous research in the area of using deep learning algorithms to forecast stock prices was focused on news headlines,company reports,and a mix of daily stock fundamentals,but few studies achieved excellent results.T...Previous research in the area of using deep learning algorithms to forecast stock prices was focused on news headlines,company reports,and a mix of daily stock fundamentals,but few studies achieved excellent results.This study uses a convolutional neural network(CNN)to predict stock prices by considering a great amount of data,consisting of financial news headlines.We call our model N-CNN to distinguish it from a CNN.The main concept is to narrow the diversity of specific stock prices as they are impacted by news headlines,then horizontally expand the news headline data to a higher level for increased reliability.This model solves the problem that the number of news stories produced by a single stock does not meet the standard of previous research.In addition,we then use the number of news headlines for every stock on the China stock exchange as input to predict the probability of the highest next day stock price fluctuations.In the second half of this paper,we compare a traditional Long Short-Term Memory(LSTM)model for daily technical indicators with an LSTM model compensated by the N-CNN model.Experiments show that the final result obtained by the compensation formula can further reduce the root-mean-square error of LSTM.展开更多
The stock market is an important economic information center.The economic benefits generated by stock price prediction have attracted much attention.Although the stock market cannot be predicted accurately,the stock m...The stock market is an important economic information center.The economic benefits generated by stock price prediction have attracted much attention.Although the stock market cannot be predicted accurately,the stock market’s prediction of the trend of stock prices helps in grasping the operation law of the stock market and the influence mechanism on the economy.The autoregressive integrated moving average(ARIMA)model is one of the most widely accepted and used time series forecasting models.Therefore,this paper first compares the return on investment(ROI)of Apple and Tesla,revealing that the ROI of Tesla is much greater than that of Apple,and subsequently focuses on ARIMA model’s prediction on the available time series data,thus concluding that the ARIMA model is better than the Naïve method in predicting the change in Tesla’s stock price trend.展开更多
An admissible manifold wavelet kernel is proposed to construct manifold wavelet support vector machine(MWSVM) for stock returns forecasting.The manifold wavelet kernel is obtained by incorporating manifold theory into...An admissible manifold wavelet kernel is proposed to construct manifold wavelet support vector machine(MWSVM) for stock returns forecasting.The manifold wavelet kernel is obtained by incorporating manifold theory into wavelet technique in support vector machine(SVM).Since manifold wavelet function can yield features that describe of the stock time series both at various locations and at varying time granularities,the MWSVM can approximate arbitrary nonlinear functions and forecast stock returns accurately.The applicability and validity of MWSVM for stock returns forecasting is confirmed through experiments on real-world stock data.展开更多
Forecasting stock prices using deep learning models suffers from pro-blems such as low accuracy,slow convergence,and complex network structures.This study developed an echo state network(ESN)model to mitigate such pro...Forecasting stock prices using deep learning models suffers from pro-blems such as low accuracy,slow convergence,and complex network structures.This study developed an echo state network(ESN)model to mitigate such pro-blems.We compared our ESN with a long short-term memory(LSTM)network by forecasting the stock data of Kweichow Moutai,a leading enterprise in China’s liquor industry.By analyzing data for 120,240,and 300 days,we generated fore-cast data for the next 40,80,and 100 days,respectively,using both ESN and LSTM.In terms of accuracy,ESN had the unique advantage of capturing non-linear data.Mean absolute error(MAE)was used to present the accuracy results.The MAEs of the data forecast by ESN were 0.024,0.024,and 0.025,which were,respectively,0.065,0.007,and 0.009 less than those of LSTM.In terms of con-vergence,ESN has a reservoir state-space structure,which makes it perform faster than other models.Root-mean-square error(RMSE)was used to present the con-vergence time.In our experiment,the RMSEs of ESN were 0.22,0.27,and 0.26,which were,respectively,0.08,0.01,and 0.12 less than those of LSTM.In terms of network structure,ESN consists only of input,reservoir,and output spaces,making it a much simpler model than the others.The proposed ESN was found to be an effective model that,compared to others,converges faster,forecasts more accurately,and builds time-series analyses more easily.展开更多
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 p...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.展开更多
This paper examines the forecasting performance of different kinds of GARCH model （GRACH, EGARCH, TARCH and APARCH） under the Normal, Student-t and Generalized error distributional assumption. We compare the effect ...This paper examines the forecasting performance of different kinds of GARCH model （GRACH, EGARCH, TARCH and APARCH） under the Normal, Student-t and Generalized error distributional assumption. We compare the effect of different distributional assumption on the GARCH models. The data we analyze are the daily stocks indexes for Shenzhen Stock Exchange （SSE） in China from April 3^rd, 1991 to April 14^th, 2005. We find that improvements of the overall estimation are achieved when asymmetric GARCH models are used with student-t distribution and generalized error distribution. Moreover, it is found that TARCH and GARCH models give better forecasting performance than EGARCH and APARCH models. In forecasting performance, the model under normal distribution gives more accurate forecasting performance than non-normal densities and generalized error distributions clearly outperform the student-t densities in case of SSE.展开更多
This study systematically reviewed the literature on using the Google Search Volume Index(GSVI)as a proxy variable for investor attention and stock market movements.We analyzed 56 academic studies published between 20...This study systematically reviewed the literature on using the Google Search Volume Index(GSVI)as a proxy variable for investor attention and stock market movements.We analyzed 56 academic studies published between 2010 and 2021 using the Web of Sciences and ScienceDirect databases.The articles were classified and synthesized based on the selection criteria for building the GSVI:keywords of the search term,market region,and frequency of the data sample.Next,we analyze the effect of returns,volatility,and trading volume on the financial variables.The main results can be summarized as follows.(1)The GSVI is positively related to volatility and trading volume regardless of the keyword,market region,or frequency used for the sample.Hence,increasing investor attention toward a specific financial term will increase volatility and trading volume.(2)The GSVI can improve forecasting models for stock market movements.To conclude,this study consolidates,for the first time,the research literature on GSVI,which is highly valuable for academic practitioners in the area.展开更多
Developing effective strategies to earn excess returns in the stock market is a cutting-edge topic in the field of economics.At the same time,stock price forecasting that supports trading strategies is considered one ...Developing effective strategies to earn excess returns in the stock market is a cutting-edge topic in the field of economics.At the same time,stock price forecasting that supports trading strategies is considered one of the most challenging tasks.Therefore,this study analyzes and extracts news media data,expert comments,social opinion data,and pandemic text data using natural language processing,and then combines the data with a deep learning model to forecast future stock price patterns based on historical stock prices.An interval constraint-based trading strategy is constructed.Using data from several typical stocks in the Chinese stock market during the COVID-19 period,the empirical studies and trading simulations show,first,that the sentiment composite index and the deep learning model can improve the accuracy of stock price forecasting.Second,the interval constraint-based trading strategy based on the proposed approach can effectively enhance returns and thus,can assist investors in decision-making.展开更多
The new energy industry is strongly supported by the state,and accurate forecasting of stock price can lead to better understanding of its development.However,factors such as cost and ease of use of new energy,as well...The new energy industry is strongly supported by the state,and accurate forecasting of stock price can lead to better understanding of its development.However,factors such as cost and ease of use of new energy,as well as economic situation and policy environment,have led to continuous changes in its stock price and increased stock price volatility.By calculating the Lyapunov index and observing the Poincarésurface of the section,we find that the sample of the China Securities Index Green Power 50 Index has chaotic characteristics,and the data indicate strong volatility and uncertainty.This study proposes a new method of stock price index prediction,namely,EWT-S-ALOSVR.Empirical wavelet decomposition extracts features from multiple factors affecting stock prices to form multiple sub-columns with features,significantly reducing the complexity of the stock price series.Support vector regression is well suited for dealing with nonlinear stock price series,and the support vector machine model parameters are selected using random wandering and picking elites via Ant Lion Optimization,making stock price prediction more accurate.展开更多
This paper proposes the generalized regression neural network(GRNN)model and multi-GRNN model with a gating network by selecting the data of Shanghai index,the stocks of Shanghai Pudong Development Bank(SPDB),Dongfeng...This paper proposes the generalized regression neural network(GRNN)model and multi-GRNN model with a gating network by selecting the data of Shanghai index,the stocks of Shanghai Pudong Development Bank(SPDB),Dongfeng Automobile and Baotou Steel.We analyze the two models using Matlab software to predict the opening price respectively.Through building a softmax excitation function,the multi-GRNN model with a gating network can obtain the best weights.Using the data of the four groups,the average of forecasting errors of 4 groups by GRNN neural model is 0.012 208,while the average of the multi-GRNN models's with a gating network is 0.002 659.Compared with the real data,it is found that the both results predicted by the two models have small mean square prediction errors.So the two models are suitable to be adopted to process a large quantity of data,furthermore the multi-GRNN model with a gating network is better than the GRNN model.展开更多
Stocks that are fundamentally connected with each other tend to move together.Considering such common trends is believed to benefit stock movement forecasting tasks.However,such signals are not trivial to model becaus...Stocks that are fundamentally connected with each other tend to move together.Considering such common trends is believed to benefit stock movement forecasting tasks.However,such signals are not trivial to model because the connections among stocks are not physically presented and need to be estimated from volatile data.Motivated by this observation,we propose a framework that incorporates the inter-connection of firms to forecast stock prices.To effectively utilize a large set of fundamental features,we further design a novel pipeline.First,we use variational autoencoder(VAE)to reduce the dimension of stock fundamental information and then cluster stocks into a graph structure(fundamentally clustering).Second,a hybrid model of graph convolutional network and long-short term memory network(GCN-LSTM)with an adjacency graph matrix(learnt from VAE)is proposed for graph-structured stock market forecasting.Experiments on minute-level U.S.stock market data demonstrate that our model effectively captures both spatial and temporal signals and achieves superior improvement over baseline methods.The proposed model is promising for other applications in which there is a possible but hidden spatial dependency to improve time-series prediction.展开更多
In stock market forecasting,the identification of critical features that affect the performance of machine learning(ML)models is crucial to achieve accurate stock price predictions.Several review papers in the literat...In stock market forecasting,the identification of critical features that affect the performance of machine learning(ML)models is crucial to achieve accurate stock price predictions.Several review papers in the literature have focused on various ML,statistical,and deep learning-based methods used in stock market forecasting.However,no survey study has explored feature selection and extraction techniques for stock market forecasting.This survey presents a detailed analysis of 32 research works that use a combination of feature study and ML approaches in various stock market applications.We conduct a systematic search for articles in the Scopus and Web of Science databases for the years 2011–2022.We review a variety of feature selection and feature extraction approaches that have been successfully applied in the stock market analyses presented in the articles.We also describe the combination of feature analysis techniques and ML methods and evaluate their performance.Moreover,we present other survey articles,stock market input and output data,and analyses based on various factors.We find that correlation criteria,random forest,principal component analysis,and autoencoder are the most widely used feature selection and extraction techniques with the best prediction accuracy for various stock market applications.展开更多
Big data analytic techniques associated with machine learning algorithms are playing an increasingly important role in various application fields,including stock market investment.However,few studies have focused on f...Big data analytic techniques associated with machine learning algorithms are playing an increasingly important role in various application fields,including stock market investment.However,few studies have focused on forecasting daily stock market returns,especially when using powerful machine learning techniques,such as deep neural networks(DNNs),to perform the analyses.DNNs employ various deep learning algorithms based on the combination of network structure,activation function,and model parameters,with their performance depending on the format of the data representation.This paper presents a comprehensive big data analytics process to predict the daily return direction of the SPDR S&P 500 ETF(ticker symbol:SPY)based on 60 financial and economic features.DNNs and traditional artificial neural networks(ANNs)are then deployed over the entire preprocessed but untransformed dataset,along with two datasets transformed via principal component analysis(PCA),to predict the daily direction of future stock market index returns.While controlling for overfitting,a pattern for the classification accuracy of the DNNs is detected and demonstrated as the number of the hidden layers increases gradually from 12 to 1000.Moreover,a set of hypothesis testing procedures are implemented on the classification,and the simulation results show that the DNNs using two PCA-represented datasets give significantly higher classification accuracy than those using the entire untransformed dataset,as well as several other hybrid machine learning algorithms.In addition,the trading strategies guided by the DNN classification process based on PCA-represented data perform slightly better than the others tested,including in a comparison against two standard benchmarks.展开更多
Accurate forecasting of changes in stock market indices can provide financial managers and individual investors with strategically valuable information.However,predicting the closing prices of stock indices remains a ...Accurate forecasting of changes in stock market indices can provide financial managers and individual investors with strategically valuable information.However,predicting the closing prices of stock indices remains a challenging task because stock price movements are characterized by high volatility and nonlinearity.This paper proposes a novel condensed polynomial neural network(CPNN)for the task of forecasting stock closing price indices.We developed a model that uses partial descriptions(PDs)and is limited to only two layers for the PNN architecture.The outputs of these PDs along with the original features are fed to a single output neuron,and the synaptic weight values and biases of the CPNN are optimized by a genetic algorithm.The proposed model was evaluated by predicting the next day’s closing price of five fast-growing stock indices:the BSE,DJIA,NASDAQ,FTSE,and TAIEX.In comparative testing,the proposed model proved its ability to provide closing price predictions with superior accuracy.Further,the Deibold-Mariano test justified the statistical significance of the model,establishing that this approach can be adopted as a competent financial forecasting tool.展开更多
Accurate prediction of stock market behavior is a challenging issue for financial forecasting.Artificial neural networks,such as multilayer perceptron have been established as better approximation and classification m...Accurate prediction of stock market behavior is a challenging issue for financial forecasting.Artificial neural networks,such as multilayer perceptron have been established as better approximation and classification models for this domain.This study proposes a chemical reaction optimization(CRO)based neuro-fuzzy network model for prediction of stock indices.The input vectors to the model are fuzzified by applying a Gaussian membership function,and each input is associated with a degree of membership to different classes.A multilayer perceptron with one hidden layer is used as the base model and CRO is used to the optimal weights and biases of this model.CRO was chosen because it requires fewer control parameters and has a faster convergence rate.Five statistical parameters are used to evaluate the performance of the model,and the model is validated by forecasting the daily closing indices for five major stock markets.The performance of the proposed model is compared with four state-of-art models that are trained similarly and was found to be superior.We conducted the Deibold-Mariano test to check the statistical significance of the proposed model,and it was found to be significant.This model can be used as a promising tool for financial forecasting.展开更多
This work aims to utilize deep learning methods CNN and LSTM to predict the adjusted close price of eight technical companies. The proposed model is a CNN-LSTM hybrid model, which combined CNN and LSTM layers in the m...This work aims to utilize deep learning methods CNN and LSTM to predict the adjusted close price of eight technical companies. The proposed model is a CNN-LSTM hybrid model, which combined CNN and LSTM layers in the model. It was compared with the single LSTM model and double LSTM model to evaluate its performance. The results showed the CNN-LSTM made a great prediction and can predict a more accurate value than the other two models, but it still can be improved to reduce overfitting problems.展开更多
Stock market is volatile and predicting stock prices is a challenging task.Stock prices are influenced by multiple factors,and prediction using only numerical or image features is ineffective.To solve this problem,we ...Stock market is volatile and predicting stock prices is a challenging task.Stock prices are influenced by multiple factors,and prediction using only numerical or image features is ineffective.To solve this problem,we propose a Hybrid Channel Stock model that incorporates multiple features of basic stock data,K-line charts and technical indicator factors for predicting the closing price of a stock on day n+1.The model combines multiple aspects of data and uses a multi-channel structure including improved CNN-TW,bidirectional LSTM and Transformer network.First,we construct the multi-channel branches of the multi-faceted feature fusion input network model;second,in this paper,we will use the concatenate method to stitch the output of each branch as the input of the rest of the network;the last layer in the network is the fully connected layer,which combines the linear activation function regression to output the predicted prices.Finally,we conducted extensive experiments on the Dow 30,SSH 50 and CSI100 indices.The experimental results show that the Hybrid Channel Stock method has the best performance with the smallest MSE,RMSE,MAE and MAPE compared with existing models.in addition,the experiments on different trading days validate the stability and effectiveness of the model,providing an important reference for investors to make stock investment decisions.展开更多
Neural network is widely used in stock price forecasting,but it lacks interpretability because of its“black box”characteristics.In this paper,L1-orthogonal regularization method is used in the GRU model.A decision t...Neural network is widely used in stock price forecasting,but it lacks interpretability because of its“black box”characteristics.In this paper,L1-orthogonal regularization method is used in the GRU model.A decision tree,GRU-DT,was conducted to represent the prediction process of a neural network,and some rule screening algorithms were proposed to find out significant rules in the prediction.In the empirical study,the data of 10 different industries in China’s CSI 300 were selected for stock price trend prediction,and extracted rules were compared and analyzed.And the method of technical index discretization was used to make rules easy for decision-making.Empirical results show that the AUC of the model is stable between 0.72 and 0.74,and the value of F1 and Accuracy are stable between 0.68 and 0.70,indicating that discretized technical indicators can predict the short-term trend of stock price effectively.And the fidelity of GRU-DT to the GRU model reaches 0.99.The prediction rules of different industries have some commonness and individuality.展开更多
Stock index forecasting has been one of the most widely investigated topics in the field of financial forecasting. Related studies typically advocate for tuning the parameters of forecasting models by minimizing learn...Stock index forecasting has been one of the most widely investigated topics in the field of financial forecasting. Related studies typically advocate for tuning the parameters of forecasting models by minimizing learning errors measured using statistical metrics such as the mean squared error or mean absolute percentage error. The authors argue that statistical metrics used to guide parameter tuning of forecasting models may not be meaningful, given the fact that the ultimate goal of forecasting is to facilitate investment decisions with expected profits in the future. The authors therefore introduce the Sharpe ratio into the process of model building and take it as the profit metric to guide parameter tuning rather than using the commonly adopted statistical metrics. The authors consider three widely used trading strategies, which include a na¨?ve strategy, a filter strategy and a dual moving average strategy, as investment scenarios. To verify the effectiveness of the proposed profit guided approach, the authors carry out simulation experiments using three global mainstream stock market indices. The results show that profit guided forecasting models are competitive, and in many cases produce significantly better performances than statistical error guided models. This implies thatprofit guided stock index forecasting is a worthwhile alternative over traditional stock index forecasting practices.展开更多
基金supported by the 2020 Hunan Natural Science Foundation Project"Research on the Key Technologies of a Personalized Learning Platform for Higher Vocational Students Based on Self-Expanding Knowledge Base and Multimodal Portraits"(2020JJ7041)partly supported by the National Natural Science Foundation of China(No.72073041).
文摘Stock market trading is an activity in which investors need fast and accurate information to make effective decisions.But the fact is that forecasting stock prices by using various models has been suffering from low accuracy,slow convergence,and complex parameters.This study aims to employ a mixed model to improve the accuracy of stock price prediction.We present how to use a random walk based on jump-diffusion,to obtain stock predictions with a good-fitting degree by adjusting different parameters.Aimed at getting better parameters and then using the time series model to predict the data,we employed the time series model to smooth the sequence utilizing logarithm and difference,which successfully resulted in drawing the auto-correlation figure and partial the auto-correlation figure.According to the comparative analysis,which focuses on checking the mean absolute error,including root mean square error and R square evaluation index,we have drawn a clear conclusion that our mixed model prediction effect is relatively good.In the context of Chinese stocks,the hybrid random walk model is very suitable for predicting stocks.It can“interpret”the randomness of stocks very well,and it also has an unparalleled prediction effect compared with other models.Based on the time series model’s application in this paper,the abovementioned series is more suitable for predicting trends.
基金This work was supported by the Natural Science Foundation of China(61572160).
文摘Previous research in the area of using deep learning algorithms to forecast stock prices was focused on news headlines,company reports,and a mix of daily stock fundamentals,but few studies achieved excellent results.This study uses a convolutional neural network(CNN)to predict stock prices by considering a great amount of data,consisting of financial news headlines.We call our model N-CNN to distinguish it from a CNN.The main concept is to narrow the diversity of specific stock prices as they are impacted by news headlines,then horizontally expand the news headline data to a higher level for increased reliability.This model solves the problem that the number of news stories produced by a single stock does not meet the standard of previous research.In addition,we then use the number of news headlines for every stock on the China stock exchange as input to predict the probability of the highest next day stock price fluctuations.In the second half of this paper,we compare a traditional Long Short-Term Memory(LSTM)model for daily technical indicators with an LSTM model compensated by the N-CNN model.Experiments show that the final result obtained by the compensation formula can further reduce the root-mean-square error of LSTM.
文摘The stock market is an important economic information center.The economic benefits generated by stock price prediction have attracted much attention.Although the stock market cannot be predicted accurately,the stock market’s prediction of the trend of stock prices helps in grasping the operation law of the stock market and the influence mechanism on the economy.The autoregressive integrated moving average(ARIMA)model is one of the most widely accepted and used time series forecasting models.Therefore,this paper first compares the return on investment(ROI)of Apple and Tesla,revealing that the ROI of Tesla is much greater than that of Apple,and subsequently focuses on ARIMA model’s prediction on the available time series data,thus concluding that the ARIMA model is better than the Naïve method in predicting the change in Tesla’s stock price trend.
基金the Hunan Natural Science Foundation(No. 09JJ3129)the Hunan Key Social Science Foundation (No. 09ZDB04)the Hunan Social Science Foundation (No. 08JD28)
文摘An admissible manifold wavelet kernel is proposed to construct manifold wavelet support vector machine(MWSVM) for stock returns forecasting.The manifold wavelet kernel is obtained by incorporating manifold theory into wavelet technique in support vector machine(SVM).Since manifold wavelet function can yield features that describe of the stock time series both at various locations and at varying time granularities,the MWSVM can approximate arbitrary nonlinear functions and forecast stock returns accurately.The applicability and validity of MWSVM for stock returns forecasting is confirmed through experiments on real-world stock data.
基金supported by the National Natural Science Foundation of China(No.72073041)Open Foundation for the University Innovation Platform in Hunan Province(No.18K103)+2 种基金2011 Collaborative Innovation Center for Development and Utilization of Finance and Economics Big Data Property,Universities of Hunan Province,Open Project(Nos.20181901CRP03,20181901CRP04,20181901CRP05)2020 Hunan Provincial Higher Education Teaching Reform Research Project(Nos.HNJG-2020-1130,HNJG-2020-1124)2020 General Project of Hunan Social Science Fund(No.20B16).
文摘Forecasting stock prices using deep learning models suffers from pro-blems such as low accuracy,slow convergence,and complex network structures.This study developed an echo state network(ESN)model to mitigate such pro-blems.We compared our ESN with a long short-term memory(LSTM)network by forecasting the stock data of Kweichow Moutai,a leading enterprise in China’s liquor industry.By analyzing data for 120,240,and 300 days,we generated fore-cast data for the next 40,80,and 100 days,respectively,using both ESN and LSTM.In terms of accuracy,ESN had the unique advantage of capturing non-linear data.Mean absolute error(MAE)was used to present the accuracy results.The MAEs of the data forecast by ESN were 0.024,0.024,and 0.025,which were,respectively,0.065,0.007,and 0.009 less than those of LSTM.In terms of con-vergence,ESN has a reservoir state-space structure,which makes it perform faster than other models.Root-mean-square error(RMSE)was used to present the con-vergence time.In our experiment,the RMSEs of ESN were 0.22,0.27,and 0.26,which were,respectively,0.08,0.01,and 0.12 less than those of LSTM.In terms of network structure,ESN consists only of input,reservoir,and output spaces,making it a much simpler model than the others.The proposed ESN was found to be an effective model that,compared to others,converges faster,forecasts more accurately,and builds time-series analyses more easily.
文摘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.
文摘This paper examines the forecasting performance of different kinds of GARCH model （GRACH, EGARCH, TARCH and APARCH） under the Normal, Student-t and Generalized error distributional assumption. We compare the effect of different distributional assumption on the GARCH models. The data we analyze are the daily stocks indexes for Shenzhen Stock Exchange （SSE） in China from April 3^rd, 1991 to April 14^th, 2005. We find that improvements of the overall estimation are achieved when asymmetric GARCH models are used with student-t distribution and generalized error distribution. Moreover, it is found that TARCH and GARCH models give better forecasting performance than EGARCH and APARCH models. In forecasting performance, the model under normal distribution gives more accurate forecasting performance than non-normal densities and generalized error distributions clearly outperform the student-t densities in case of SSE.
文摘This study systematically reviewed the literature on using the Google Search Volume Index(GSVI)as a proxy variable for investor attention and stock market movements.We analyzed 56 academic studies published between 2010 and 2021 using the Web of Sciences and ScienceDirect databases.The articles were classified and synthesized based on the selection criteria for building the GSVI:keywords of the search term,market region,and frequency of the data sample.Next,we analyze the effect of returns,volatility,and trading volume on the financial variables.The main results can be summarized as follows.(1)The GSVI is positively related to volatility and trading volume regardless of the keyword,market region,or frequency used for the sample.Hence,increasing investor attention toward a specific financial term will increase volatility and trading volume.(2)The GSVI can improve forecasting models for stock market movements.To conclude,this study consolidates,for the first time,the research literature on GSVI,which is highly valuable for academic practitioners in the area.
基金partly supported by the National Natural Science Foundation of China under Grants No.72171223,No.71801213,and No.71988101the National Key R&D Program of China under Grants No.2021ZD0111204。
文摘Developing effective strategies to earn excess returns in the stock market is a cutting-edge topic in the field of economics.At the same time,stock price forecasting that supports trading strategies is considered one of the most challenging tasks.Therefore,this study analyzes and extracts news media data,expert comments,social opinion data,and pandemic text data using natural language processing,and then combines the data with a deep learning model to forecast future stock price patterns based on historical stock prices.An interval constraint-based trading strategy is constructed.Using data from several typical stocks in the Chinese stock market during the COVID-19 period,the empirical studies and trading simulations show,first,that the sentiment composite index and the deep learning model can improve the accuracy of stock price forecasting.Second,the interval constraint-based trading strategy based on the proposed approach can effectively enhance returns and thus,can assist investors in decision-making.
基金Key Research Project in Universities of Henan Province(No.24B480012)Science and Technology of Henan Province of China(No.182400410419)+1 种基金the Foundation for Fostering the National Foundation of Pingdingshan University(No.PXY-PYJJ-2016006)National Science and Technology Council,Taiwan(MOST 111-2410-H-161-001).
文摘The new energy industry is strongly supported by the state,and accurate forecasting of stock price can lead to better understanding of its development.However,factors such as cost and ease of use of new energy,as well as economic situation and policy environment,have led to continuous changes in its stock price and increased stock price volatility.By calculating the Lyapunov index and observing the Poincarésurface of the section,we find that the sample of the China Securities Index Green Power 50 Index has chaotic characteristics,and the data indicate strong volatility and uncertainty.This study proposes a new method of stock price index prediction,namely,EWT-S-ALOSVR.Empirical wavelet decomposition extracts features from multiple factors affecting stock prices to form multiple sub-columns with features,significantly reducing the complexity of the stock price series.Support vector regression is well suited for dealing with nonlinear stock price series,and the support vector machine model parameters are selected using random wandering and picking elites via Ant Lion Optimization,making stock price prediction more accurate.
基金Postdoctoral Granted Financial Support from China Postdoctoral Science Foundation(20100481307)Natural Science Foundation of Shanxi Province,China(No.2009011018-3)
文摘This paper proposes the generalized regression neural network(GRNN)model and multi-GRNN model with a gating network by selecting the data of Shanghai index,the stocks of Shanghai Pudong Development Bank(SPDB),Dongfeng Automobile and Baotou Steel.We analyze the two models using Matlab software to predict the opening price respectively.Through building a softmax excitation function,the multi-GRNN model with a gating network can obtain the best weights.Using the data of the four groups,the average of forecasting errors of 4 groups by GRNN neural model is 0.012 208,while the average of the multi-GRNN models's with a gating network is 0.002 659.Compared with the real data,it is found that the both results predicted by the two models have small mean square prediction errors.So the two models are suitable to be adopted to process a large quantity of data,furthermore the multi-GRNN model with a gating network is better than the GRNN model.
文摘Stocks that are fundamentally connected with each other tend to move together.Considering such common trends is believed to benefit stock movement forecasting tasks.However,such signals are not trivial to model because the connections among stocks are not physically presented and need to be estimated from volatile data.Motivated by this observation,we propose a framework that incorporates the inter-connection of firms to forecast stock prices.To effectively utilize a large set of fundamental features,we further design a novel pipeline.First,we use variational autoencoder(VAE)to reduce the dimension of stock fundamental information and then cluster stocks into a graph structure(fundamentally clustering).Second,a hybrid model of graph convolutional network and long-short term memory network(GCN-LSTM)with an adjacency graph matrix(learnt from VAE)is proposed for graph-structured stock market forecasting.Experiments on minute-level U.S.stock market data demonstrate that our model effectively captures both spatial and temporal signals and achieves superior improvement over baseline methods.The proposed model is promising for other applications in which there is a possible but hidden spatial dependency to improve time-series prediction.
基金funded by The University of Groningen and Prospect Burma organization.
文摘In stock market forecasting,the identification of critical features that affect the performance of machine learning(ML)models is crucial to achieve accurate stock price predictions.Several review papers in the literature have focused on various ML,statistical,and deep learning-based methods used in stock market forecasting.However,no survey study has explored feature selection and extraction techniques for stock market forecasting.This survey presents a detailed analysis of 32 research works that use a combination of feature study and ML approaches in various stock market applications.We conduct a systematic search for articles in the Scopus and Web of Science databases for the years 2011–2022.We review a variety of feature selection and feature extraction approaches that have been successfully applied in the stock market analyses presented in the articles.We also describe the combination of feature analysis techniques and ML methods and evaluate their performance.Moreover,we present other survey articles,stock market input and output data,and analyses based on various factors.We find that correlation criteria,random forest,principal component analysis,and autoencoder are the most widely used feature selection and extraction techniques with the best prediction accuracy for various stock market applications.
文摘Big data analytic techniques associated with machine learning algorithms are playing an increasingly important role in various application fields,including stock market investment.However,few studies have focused on forecasting daily stock market returns,especially when using powerful machine learning techniques,such as deep neural networks(DNNs),to perform the analyses.DNNs employ various deep learning algorithms based on the combination of network structure,activation function,and model parameters,with their performance depending on the format of the data representation.This paper presents a comprehensive big data analytics process to predict the daily return direction of the SPDR S&P 500 ETF(ticker symbol:SPY)based on 60 financial and economic features.DNNs and traditional artificial neural networks(ANNs)are then deployed over the entire preprocessed but untransformed dataset,along with two datasets transformed via principal component analysis(PCA),to predict the daily direction of future stock market index returns.While controlling for overfitting,a pattern for the classification accuracy of the DNNs is detected and demonstrated as the number of the hidden layers increases gradually from 12 to 1000.Moreover,a set of hypothesis testing procedures are implemented on the classification,and the simulation results show that the DNNs using two PCA-represented datasets give significantly higher classification accuracy than those using the entire untransformed dataset,as well as several other hybrid machine learning algorithms.In addition,the trading strategies guided by the DNN classification process based on PCA-represented data perform slightly better than the others tested,including in a comparison against two standard benchmarks.
文摘Accurate forecasting of changes in stock market indices can provide financial managers and individual investors with strategically valuable information.However,predicting the closing prices of stock indices remains a challenging task because stock price movements are characterized by high volatility and nonlinearity.This paper proposes a novel condensed polynomial neural network(CPNN)for the task of forecasting stock closing price indices.We developed a model that uses partial descriptions(PDs)and is limited to only two layers for the PNN architecture.The outputs of these PDs along with the original features are fed to a single output neuron,and the synaptic weight values and biases of the CPNN are optimized by a genetic algorithm.The proposed model was evaluated by predicting the next day’s closing price of five fast-growing stock indices:the BSE,DJIA,NASDAQ,FTSE,and TAIEX.In comparative testing,the proposed model proved its ability to provide closing price predictions with superior accuracy.Further,the Deibold-Mariano test justified the statistical significance of the model,establishing that this approach can be adopted as a competent financial forecasting tool.
文摘Accurate prediction of stock market behavior is a challenging issue for financial forecasting.Artificial neural networks,such as multilayer perceptron have been established as better approximation and classification models for this domain.This study proposes a chemical reaction optimization(CRO)based neuro-fuzzy network model for prediction of stock indices.The input vectors to the model are fuzzified by applying a Gaussian membership function,and each input is associated with a degree of membership to different classes.A multilayer perceptron with one hidden layer is used as the base model and CRO is used to the optimal weights and biases of this model.CRO was chosen because it requires fewer control parameters and has a faster convergence rate.Five statistical parameters are used to evaluate the performance of the model,and the model is validated by forecasting the daily closing indices for five major stock markets.The performance of the proposed model is compared with four state-of-art models that are trained similarly and was found to be superior.We conducted the Deibold-Mariano test to check the statistical significance of the proposed model,and it was found to be significant.This model can be used as a promising tool for financial forecasting.
文摘This work aims to utilize deep learning methods CNN and LSTM to predict the adjusted close price of eight technical companies. The proposed model is a CNN-LSTM hybrid model, which combined CNN and LSTM layers in the model. It was compared with the single LSTM model and double LSTM model to evaluate its performance. The results showed the CNN-LSTM made a great prediction and can predict a more accurate value than the other two models, but it still can be improved to reduce overfitting problems.
基金supported by these three foundation programs:the Science and Technology Research Project(Youth)of Chongqing Municipal Education Commission(KJQN202201142)the Chongqing Research Program of Basic Research Frontier Technology(CSTB2022BSXM-JCX0069CCCC)the Training Program of the National Natural Science Foundation of China and National Social Science Fund of China of Chongqing University of Technology(2022PYZ030)。
文摘Stock market is volatile and predicting stock prices is a challenging task.Stock prices are influenced by multiple factors,and prediction using only numerical or image features is ineffective.To solve this problem,we propose a Hybrid Channel Stock model that incorporates multiple features of basic stock data,K-line charts and technical indicator factors for predicting the closing price of a stock on day n+1.The model combines multiple aspects of data and uses a multi-channel structure including improved CNN-TW,bidirectional LSTM and Transformer network.First,we construct the multi-channel branches of the multi-faceted feature fusion input network model;second,in this paper,we will use the concatenate method to stitch the output of each branch as the input of the rest of the network;the last layer in the network is the fully connected layer,which combines the linear activation function regression to output the predicted prices.Finally,we conducted extensive experiments on the Dow 30,SSH 50 and CSI100 indices.The experimental results show that the Hybrid Channel Stock method has the best performance with the smallest MSE,RMSE,MAE and MAPE compared with existing models.in addition,the experiments on different trading days validate the stability and effectiveness of the model,providing an important reference for investors to make stock investment decisions.
基金National Defense Science and Technology Innovation Special ZoneProject (No. 18-163-11-ZT-002-045-04).
文摘Neural network is widely used in stock price forecasting,but it lacks interpretability because of its“black box”characteristics.In this paper,L1-orthogonal regularization method is used in the GRU model.A decision tree,GRU-DT,was conducted to represent the prediction process of a neural network,and some rule screening algorithms were proposed to find out significant rules in the prediction.In the empirical study,the data of 10 different industries in China’s CSI 300 were selected for stock price trend prediction,and extracted rules were compared and analyzed.And the method of technical index discretization was used to make rules easy for decision-making.Empirical results show that the AUC of the model is stable between 0.72 and 0.74,and the value of F1 and Accuracy are stable between 0.68 and 0.70,indicating that discretized technical indicators can predict the short-term trend of stock price effectively.And the fidelity of GRU-DT to the GRU model reaches 0.99.The prediction rules of different industries have some commonness and individuality.
基金supported by the Natural Science Foundation of China under Grant Nos.71601147,71571080,and 71501079the Central Universities under Grant No.104-413000017the China Postdoctoral Science Foundation under Grant No.2015M582280
文摘Stock index forecasting has been one of the most widely investigated topics in the field of financial forecasting. Related studies typically advocate for tuning the parameters of forecasting models by minimizing learning errors measured using statistical metrics such as the mean squared error or mean absolute percentage error. The authors argue that statistical metrics used to guide parameter tuning of forecasting models may not be meaningful, given the fact that the ultimate goal of forecasting is to facilitate investment decisions with expected profits in the future. The authors therefore introduce the Sharpe ratio into the process of model building and take it as the profit metric to guide parameter tuning rather than using the commonly adopted statistical metrics. The authors consider three widely used trading strategies, which include a na¨?ve strategy, a filter strategy and a dual moving average strategy, as investment scenarios. To verify the effectiveness of the proposed profit guided approach, the authors carry out simulation experiments using three global mainstream stock market indices. The results show that profit guided forecasting models are competitive, and in many cases produce significantly better performances than statistical error guided models. This implies thatprofit guided stock index forecasting is a worthwhile alternative over traditional stock index forecasting practices.
