Scale-Free Behavior in Weighted Stock Network

A weighted stock network model of stock market is presented based on the complex network theory. The model is a weighted random network, in which each vertex denotes a stock, and the weight assigned to each edge is the cross-correlation coefficient of returns. Analysis of A shares listed at Shanghai Stock Exchange finds that the influence-strength （IS） follows a power-law distribution with the exponent of 2.58. The empirical analysis results show that there are a few stocks whose price fluctuations can powerfully influence the price dynamics of other stocks in the same market. Further econometric analysis reveals that there are significant differences between the positive IS and the negative IS.

Stock Price Prediction Based on the Bi-GRU-Attention Model

The stock market, as one of the hotspots in the financial field, forms a data system with a huge volume of data and complex relationships between various factors, making stock price prediction an area of keen interest for further in-depth mining and research. Mathematical statistics methods struggle to deal with nonlinear relationships in practical applications, making it difficult to explore deep information about stocks. Meanwhile, machine learning methods, particularly neural network models and composite models, which have achieved outstanding results in other fields, are being applied to the stock market with significant results. However, researchers have found that these methods do not grasp the essential information of the data as well as expected. In response to these issues, researchers are exploring better neural network models and combining them with other methods to analyze stock data. Thus, this paper proposes the ABiGRU composite model, which combines the attention mechanism and bidirectional gated recurrent unit (GRU) that can effectively extract data features for stock price prediction research. Models such as LSTM, GRU, and Bi-LSTM are selected for comparative experiments. To ensure the credibility and representativeness of the research data, daily stock price indices of BYD are chosen for closing price prediction studies across different models. The results show that the ABiGRU model has a lower prediction error and better fitting effect on three index-based stock prices, enhancing the learning efficiency of the neural network model and demonstrating good prediction stability. This suggests that the ABiGRU model is highly adaptable for stock price prediction.

General election effect on the network topology of Pakistan’s stock market: network-based study of a political event

To examine the interdependency and evolution of Pakistan’s stock market,we consider the cross-correlation coefficients of daily stock returns belonging to the blue chip Karachi stock exchange(KSE-100)index.Using the minimum spanning tree network-based method,we extend the financial network literature by examining the topological properties of the network and generating six minimum spanning tree networks around three general elections in Pakistan.Our results reveal a star-like structure after the general elections of 2018 and before those in 2008,and a tree-like structure otherwise.We also highlight key nodes,the presence of different clusters,and compare the differences between the three elections.Additionally,the sectorial centrality measures reveal economic expansion in three industrial sectors—cement,oil and gas,and fertilizers.Moreover,a strong overall intermediary role of the fertilizer sector is observed.The results indicate a structural change in the stock market network due to general elections.Consequently,through this analysis,policy makers can focus on monitoring key nodes around general elections to estimate stock market stability,while local and international investors can form optimal diversification strategies.

Predicting Stock Movement Using Sentiment Analysis of Twitter Feed with Neural Networks

External factors, such as social media and financial news, can have wide-spread effects on stock price movement. For this reason, social media is considered a useful resource for precise market predictions. In this paper, we show the effectiveness of using Twitter posts to predict stock prices. We start by training various models on the Sentiment 140 Twitter data. We found that Support Vector Machines (SVM) performed best (0.83 accuracy) in the sentimental analysis, so we used it to predict the average sentiment of tweets for each day that the market was open. Next, we use the sentimental analysis of one year’s data of tweets that contain the “stock market”, “stocktwits”, “AAPL” keywords, with the goal of predicting the corresponding stock prices of Apple Inc. (AAPL) and the US’s Dow Jones Industrial Average (DJIA) index prices. Two models, Boosted Regression Trees and Multilayer Perceptron Neural Networks were used to predict the closing price difference of AAPL and DJIA prices. We show that neural networks perform substantially better than traditional models for stocks’ price prediction.

Application of multi-GRNN with a gating network in stock prices forecast

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.

Theoretical analyses of stock correlations affected by subprime crisis and total assets: Network properties and corresponding physical mechanisms

In the field of statistical mechanics and system science, it is acknowledged that the financial crisis has a profound influence on stock market. However, the influence of total asset of enterprise on stock quote was not considered in the previous studies. In this work, a modified cross-correlation matrix that focuses on the influence of total asset on stock quote is introduced into the analysis of the stocks collected from Asian and American stock markets, which is different from the previous studies. The key results are obtained as follows. Firstly, stock is more greatly correlated with big asset than with small asset. Secondly, the higher the correlation coefficient among stocks, the larger the eigenvector is. Thirdly, in different periods, like the pre-subprime crisis period and the peak of subprime crisis period, Asian stock quotes show that the component of the third eigenvector of the cross-correlation matrix decreases with the asset of the enterprise decreasing.Fourthly, by simulating the threshold network, the small network constructed by 10 stocks with large assets can show the large network state constructed by 30 stocks. In this research we intend to fully explain the physical mechanism for understanding the historical correlation between stocks and provide risk control strategies in the future.

Hot Events Detection of Stock Market Based on Time Series Data of Stock and Text Data of Network Public Opinion

With the highly integration of the Internet world and the real world, Internet information not only provides real-time and effective data for financial investors, but also helps them understand market dynamics, and enables investors to quickly identify relevant financial events that may lead to stock market volatility. However, in the research of event detection in the financial field, many studies are focused on micro-blog, news and other network text information. Few scholars have studied the characteristics of financial time series data. Considering that in the financial field, the occurrence of an event often affects both the online public opinion space and the real transaction space, so this paper proposes a multi-source heterogeneous information detection method based on stock transaction time series data and online public opinion text data to detect hot events in the stock market. This method uses outlier detection algorithm to extract the time of hot events in stock market based on multi-member fusion. And according to the weight calculation formula of the feature item proposed in this paper, this method calculates the keyword weight of network public opinion information to obtain the core content of hot events in the stock market. Finally, accurate detection of stock market hot events is achieved.

Prediction of the Bombay Stock Exchange (BSE) Market Returns Using Artificial Neural Network and Genetic Algorithm

Stock Market is the market for security where organized issuance and trading of Stocks take place either through exchange or over the counter in electronic or physical form. It plays an important role in canalizing capital from the investors to the business houses, which consequently leads to the availability of funds for business expansion. In this paper, we investigate to predict the daily excess returns of Bombay Stock Exchange (BSE) indices over the respective Treasury bill rate returns. Initially, we prove that the excess return time series do not fluctuate randomly. We are applying the prediction models of Autoregressive feed forward Artificial Neural Networks (ANN) to predict the excess return time series using lagged value. For the Artificial Neural Networks model using a Genetic Algorithm is constructed to choose the optimal topology. This paper examines the feasibility of the prediction task and provides evidence that the markets are not fluctuating randomly and finally, to apply the most suitable prediction model and measure their efficiency.

基于动态异构网络的股价预测

股票预测通常被形式化为非线性的时间序列预测任务,但很少有研究者试图通过技术面数据去系统地揭示股票市场内在结构,例如股票上涨或下跌背后的原因可能是业务领域之间的合作或冲突,这些额外信息的增加有助于判断股票的未来趋势。为了充分真实刻画股票市场的交易状态,表达股票之间显式或隐式的关系,提出一种基于动态异构网络的股价预测模型sDHN(stock dynamic heterogeneous network),综合股票以及所属行业和地域,将其建模为动态异构网络。该模型在网络上引入动态时序特征,创新融合股票节点的四种不同技术层面的相似性图,生成富信息异构图,最后聚合不同元路径中隐含的语义信息生成嵌入,从异构图的角度充分探索股票之间的潜在关联。此外,在三个真实世界的股票数据集上进行了大量实验,所提出的模型准确率比所有基线模型均高出5%~34%,F_(1)-score则高出11.5%~37%,并且在图解释上证明了该方法的有效性。

序列稀疏自回归方法及其在美股做空数据分析上的应用

采用序列稀疏回归的思路来处理向量自回归模型,并设计适用于大规模时间序列数据分析的序列稀疏自回归方法。研究表明:从因子角度刻画向量自回归模型可以有效地将稀疏矩阵估计问题分解成稀疏奇异向量的估计问题,从而极大地提高了计算效率。以1523家美股上市公司1973年1月—2014年12月的做空数据为例,利用此方法探索公司之间的大规模做空关联网络。研究发现:此方法可以有效地恢复股票做空份额(即某一公司的空头股份数量)与股票收益率之间隐藏的关联网络,对于股票风险溢价研究具有一定启发意义。

基于定性网络模型评价生态调控情景对海洋牧场生态系统的影响——以獐子岛海洋牧场为例

在海洋牧场建设过程中,生态调控方式实施后对海洋牧场生态系统的影响通常难以预测,这对海洋牧场的生态安全和高质量发展提出了严峻挑战。为此,建立了一种基于定性网络模型(Qualitative network model,QNM)的海洋牧场生态系统模拟评价方法,并以獐子岛海洋牧场近岸增殖海域作为研究区域,构建以增殖目标种为核心的定性网络模型,模拟评估海洋牧场3种不同类型的生态调控情景(增殖目标种、移除捕食者、海藻场修复)及其复合条件下,牧场群落范围内的响应,分析海洋牧场生态调控策略与生物功能群变化之间潜在关系。结果显示:目标种增殖(仿刺参和虾夷扇贝)产生的上行效应导致其捕食者呈现积极响应,产生的下行效应导致其它底栖动物、浮游植物和有机碎屑等功能群呈现消极响应,移除捕食者海星产生的下行效应导致虾夷扇贝呈现积极响应,表明在增殖区清除敌害生物的重要性,海藻场修复对整个群落有明显的积极影响,体现了海藻场在养护近岸生态系统的重要生态意义。研究表明:QNM可有效识别生态系统潜在的营养级联效应,评估生物功能群的响应,基于QNM的海洋牧场生态调控模拟评价方法,突破了定量食物网模型在数据有限系统中使用的局限性,可为海洋牧场建设的生态调控策略制定提供科学参考。

机构投资者博彩行为的传染——基金竞争网络与彩票型股票资产配置

本文以我国2005—2021年开放式股票型与混合型基金重仓持股数据为研究样本,通过三维空间模型构建基金竞争网络,实证检验了基金竞争网络对基金经理彩票型股票资产配置的影响。研究发现,同业基金在彩票型股票资产配置上呈现出显著的一致性,意味着基金经理倾向于学习与模仿同业基金的彩票型股票资产配置策略。上述结论在考虑了外围基金的持仓变动、市场信息等因素的影响后依旧成立。潜在机制表明,在竞争激烈的环境中,基金经理面临更大的解雇风险,促使其博彩偏好上升,从而趋向于学习与模仿同业基金的彩票型资产配置策略。异质性分析表明,同业基金对彩票型资产配置策略的模仿行为存在“家族效应”与“同城效应”。该模仿行为与基金经理的性别、学历等个人特征相关,且在调仓情景下更显著。拓展性分析发现,同业竞争引致的彩票型资产配置可能进一步引发股价崩盘风险。本文为监管层理解机构投资者的博彩行为、制定合理的市场竞争机制提供了有益参考。

基于动态选择预测器的深度强化学习投资组合模型

近年来,投资组合管理问题在人工智能领域得到了广泛的研究,但现有的基于深度学习的量化交易方法还存在一些问题。首先,对股票的预测模式单一,通常一个模型只能训练出一个交易专家,交易决策也仅根据模型预测结果作出;其次,模型使用的数据源相对单一,只考虑了股票自身数据,忽略了整个市场风险对股票的影响。针对上述问题,提出了基于动态选择预测器的强化学习模型(DSDRL)。该模型分为3部分,首先提取股票数据的特征并传入多个预测器中,针对不同的投资策略训练多个预测模型,用动态选择器得到当前最优预测结果;其次,利用市场环境评价模块对当前市场风险进行量化,得到合适的投资金额比例;最后,在前两个模块的基础上建立了一种深度强化学习模型模拟真实的交易环境,基于预测的结果和投资金额比例得到实际投资组合策略。文中使用中证500和标普500的日k线数据进行测试验证,结果表明,此模型在夏普率等指标上均优于其他参照模型。

多层网络视角下沪深港股票市场关联性演化研究

随着“沪港通”和“深港通”的开通,沪深港股票市场间关联性日趋紧密,而充分认识沪深港股票市场间关联性的演化特征,对维持沪深港股票市场的稳定具有重要的价值.本文从多层网络视角出发,同时考虑股票收益率间的相关性和投资者情绪间的相关性,构建沪深港股票市场多层网络,进而探究沪深港股票市场间的关联关系及演化过程.结果表明,“沪港通”开通后,沪深港股票市场收益率间的关联性以及投资者情绪间的关联性没有显著增强,但不同市场间收益率与投资者情绪的关联性显著增强.而“深港通”开通后,沪深港股票市场收益率间、投资者情绪间以及收益率与投资者情绪间的关联性均显著增强.此外,不同市场之间收益率与投资者情绪间的交互关系并不对称,沪深两市股票收益率与港市股票投资者情绪的关联性较高,而港市股票收益率与沪深两市股票投资者情绪的关联性却较低.

贸易网络中心性对国际证券资本流动的影响研究——基于全球市场的经验数据

基于1990—2019年全球131个国家(地区)的进出口贸易等数据构建贸易网络中心性测度指标,采用面板回归模型实证检验贸易网络中心性对国际证券资本流动的影响效应。结果表明:贸易网络中心性水平提升,该国国际证券资本(净)流入将受到抑制;抑制作用的机制通过影响股票和债券的国际投资得以实现,次贷危机期间及发达国家所受到的抑制效果更趋明显;抑制作用的强弱受各国的股市波动率、银行业发展水平和贷款风险溢价等因素影响,套利因素将促使资本经由处于贸易网络中心的国家流到边缘国家。我国应结合贸易网络中心优势产生的外部影响力和控制力,优化政策体系,提高国际证券资本配置效率。

北方森林乔木层碳储量的估计及空间分析

利用遥感的方式对森林乔木层碳储量(Aboveground Biomass Carbon Stocks,ABGCS)以及乔木层碳储量的光饱和值进行精准估测,以期替代传统大面积调查的繁琐工序,为碳储量的估测提供参考和依据,提高森林可持续经营管理的效率。以2017年黑龙江省伊春市嘉荫县森林乔木层碳储量(ABGCS)作为研究对象,利用Landsat8 OLI遥感影像以及森林资源二类调查数据,构建参数模型多元逐步回归模型(Stepwise Multiple-Regression,SMR),非参数模型BP神经网络模型(BP neural network,BP-NN)、随机森林模型(Random Forest,RF)、支持向量回归模型(Support Vector Machine,SVR)对嘉荫县地区ABGCS进行估测和反演其空间分布情况。研究结果表明,非参数模型的估测精度明显高于参数模型,其中3种非参数模型(BP-NN、RF、SVR)相较于参数模型(SMR),拟合精度分别提高了25.0%、12.2%、7.3%;综合比较4种模型十折交叉验证的评价指标,分析得出模型性能优劣为BP-NN>RF>SVR>SMR,其中BP-NN模型拟合出最大的决定系数(R2为0.785)和最小的均方根误差(RMSE为3.572 t/hm2)、均方误差(MSE为12.757 t/hm2)、平均绝对误差(MAE为2.687 t/hm2);从碳储量残差分段检验结果来看,4种模型均存在碳储量不同程度上高值低估和低值高估的情况,BP-NN模型在各碳储量分段的平均残差(ME)和相对平均残差(MRE)值均为最小,其泛化能力较强;利用立方项模型确定ABGCS的光饱和值为63.056 t/hm2,与BP-NN所预测的ABGCS光饱和值接近(64.232 t/hm2)。因此,BP-NN模型在估测嘉荫县ABGCS具有较为理想的效果,为森林碳储量动态监测及研究提供重要依据。

灰狼算法优化BP神经网络的股价预测

探讨使用灰狼算法改进BP神经网络的方法,旨在提高BP神经网络的训练效果和性能。首先,介绍了BP神经网络的基本原理和灰狼算法的基本概念。然后,将灰狼算法应用于BP神经网络的权重和偏置值的优化过程中,通过调整这些参数来降低误差函数,从而提高网络的准确性和收敛速度。实验结果表明:灰狼算法优化的BP神经网络具有较好的性能和泛化能力。其次,还用股票数据进行了实证分析,该模型在股票价格预测方面具有较高的准确性和稳定性,可为投资者提供有效的决策参考。最后,总结了本研究的贡献和未来的研究方向。

融合情感分析和GAN-TrellisNet的股价预测方法

将时序深度神经网络应用于股票价格预测,已成为量化金融领域的重要研究方向。时序神经网络具有很好的序列数据捕捉能力和学习记忆能力,在股票预测上有一定适用性。但是现有的模型大多存在预测准确度不高、模型结构复杂导致训练时间较长等问题.为了解决以上问题,提出了一种基于情感分析和GAN-TrellisNet的股价预测方法。提出了一个基于LSTM-CNN的情感分析模型,用于分析爬虫获取的主流金融论坛股票评论,并获得股票情感指数。为了提高预测准确度,将情感指数和百度搜索指数加入股票交易数据中作为训练集,提出了一个基于TrellisNet和CNN的改进型GAN股价预测模型,利用TrellisNet生成器的卷积特性来捕捉数据的局部特征,选取特征提取能力较强的CNN作为判别器来区别预测结果和真实股价。通过选取10只代表性股票和三种大盘指数的不同时段数据进行算法验证,结果表明,与ConvLSTM和GAN-LSTM预测模型相比,GAN-TrellisNet模型能有效缩短训练时间,提高预测准确率。