Structured Multi-Head Attention Stock Index Prediction Method Based Adaptive Public Opinion Sentiment Vector

The present study examines the impact of short-term public opinion sentiment on the secondary market,with a focus on the potential for such sentiment to cause dramatic stock price fluctuations and increase investment risk.The quantification of investment sentiment indicators and the persistent analysis of their impact has been a complex and significant area of research.In this paper,a structured multi-head attention stock index prediction method based adaptive public opinion sentiment vector is proposed.The proposedmethod utilizes an innovative approach to transform numerous investor comments on social platforms over time into public opinion sentiment vectors expressing complex sentiments.It then analyzes the continuous impact of these vectors on the market through the use of aggregating techniques and public opinion data via a structured multi-head attention mechanism.The experimental results demonstrate that the public opinion sentiment vector can provide more comprehensive feedback on market sentiment than traditional sentiment polarity analysis.Furthermore,the multi-head attention mechanism is shown to improve prediction accuracy through attention convergence on each type of input information separately.Themean absolute percentage error(MAPE)of the proposedmethod is 0.463%,a reduction of 0.294% compared to the benchmark attention algorithm.Additionally,the market backtesting results indicate that the return was 24.560%,an improvement of 8.202% compared to the benchmark algorithm.These results suggest that themarket trading strategy based on thismethod has the potential to improve trading profits.

Risk management of stock index futures

The Hong Kong Hang Seng index futures is taken as a study object and a method of empirical analysis is adopted in order to verify the validity of the application of the value-at-risk (VaR) method in the risk measurement of the stock index futures market. The results suggest that under normal market conditions it is feasible to apply the VaR method in the measurement of the market risks of stock index futures. The daily VaR value of the stock index futures provides a foreseeable profit and loss of the stock index futures. Financial supervisors can adjust their supervising strategies according to the daily VaR value. The speculators can adjust risk capital reserve rates in the same way. The application of this method in China's stock index futures market requires the solutions to specific problems: the absence of historical data, the difficult confirmation of non-risk interest rates etc.

Cost-benefit analysis of trading strategies in the stock index futures market

With the introduction of many derivatives into the capital market,including stock index futures,the trading strategies in financial markets have been gradually enriched.However,there is still no theoretical model that can determine whether these strategies are effective,what the risks are,and how costly the strategies are.We built an agent-based cross-market platform that includes five stocks and one stock index future,and constructed an evaluation system for stock index futures trading strategies.The evaluation system includes four dimensions:effectiveness,risk,occupation of capital,and impact cost.The results show that the informed strategy performs well in all aspects.The risk of the technical strategy is relatively higher than that of the other strategies.Moreover,occupation of capital and impact cost are both higher for the arbitrage strategy.Finally,the wealth of noise traders is almost lost.

Optimal Hedging Strategies of Stock Index Futures Based on the Perspective of Information Asymmetry

Based on two different risk measurement criteria, this article studied the optimal hedging strategies of stock index futures in the case of asymmetric information, and discussed the influence of insider information on the hedging effect. Through simulation analysis, it can be shown that hedging people with insider information can save hedging costs to a certain extent, which also explains the reason why investors try to obtain corporate information in actual investment activities.

The Different Performance of the Stock Market Indexes of the Three Countries in Different International Events

During the period of different changes in the global situation,the stock indexes of China,the United States,and the United Kingdom all showed different trends.Overall,during the outbreak of the epidemic,they all received a huge impact,and due to the different policies and coping strategies of various countries,the follow-up performance also varies greatly.Brexit has only had a slight impact on the British domestic market in a short period time,and China and the United States have prepared for investment in the new market after Brexit,which has also caused the corresponding market index to perform better before the follow-up.Due to the differences in the main market targets and the differences in the geographical location of countries,the negative impact on the British market was more obvious during the Russia-Ukraine conflict,while the stock indexes of China and the United States were relatively stable and even showed an upward trend.It can be seen from the data analysis that the markets in different countries are affected by time differently.With the growing correlation between the markets of various countries,investors should pay more attention to the global situation and the policy orientation of different countries.Considering risk diversification while taking policy dividends helps to obtain stable returns.

Stock index adjustments, analyst coverage and institutional holdings: Evidence from China

Using 231 pairs of matched firms from 2009 to 2012 in Chinese stock market,we find that the stock index adjustment significantly affects the analyst coverage, which in addition to the stock index leads to more analyst coverage, while deletion from the stock index has no significant effect, indicating that stock index adjustment can significantly change the information environments of firms that are added to the index. An index adjustment also affects institutional holdings in consideration of new information(e.g., changes in fundamentals and information environments). Changes in institutional holdings are partially due to changes in analyst coverage, and both index funds and other types can change their portfolios in response to changes in the target firms' informativeness.

Profit Guided or Statistical Error Guided? A Study of Stock Index Forecasting Using Support Vector Regression

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.

Stock index adjustments and analysts' forecast optimism:A quasi-natural experiment on the CSI 300 Index

As stock index adjustments comprise a basic system of capital market,their potential influence on analysts’earnings forecasts is worthy of research.Based on a research sample of 23 adjustments to the CSI 300 Index from June 2007 to June 2018 and the backup stocks announced during the same period,this study examines the impact of additions to stock index on analysts’forecast optimism using a staggered difference-in-differences model.The research results show that after stocks are added to the stock index,analysts’earnings forecast optimism about these stocks increases significantly.Cross-sectional analysis indicates that this increase is more significant when the market is bullish,institutional ownership is low,the ratio of listed brokerage firms is low,star analyst coverage is low,firms show seasoned equity offering activity,the ratio of analysts from the top five brokerage firms ranked by commission income is high,and the analysts’brokerage firms are shareholders.However,analystlevel tests find that analysts’ability helps to reduce the impact of additions to stock index on earnings forecast optimism.Furthermore,additions to stock index significantly increase analyst coverage and forecast divergence.Economic consequences tests find additions to stock index significantly increases stock price synchronization,which is partly mediated by analysts’earnings forecast optimism.This study enriches the literature on the impact of basic capital market systems and analyst behavior.The findings suggest that investors should rationally evaluate analysts’earnings forecasts for stocks added to the stock index and obtain further information from various channels to improve asset allocation efficiency.

存量地下空间更新价值评估体系研究

存量地下空间在补充地面空间资源、完善城市设施方面具有巨大的潜力,但在发展过程中凸显出功能集聚但空间匮乏、品质参差且风貌缺失、功能错位且动迁成本高、停车供给不足又缺乏联动等众多问题。因其自身特点复杂且受到的限制较多,如对于公众开放性不足,涉及多个发展目标,影响方案决策的因素较多,以及配套管理机制滞后等,加之缺乏高效的价值评估方法,导致在对其进行更新再利用时面临利用方式的单一和利用效率的不足。本文从全要素研究分析的角度出发,以现状质量、更新需求、成本投入和可获效益4个需求准则进行价值评估指标的选取,经过对多种评估方法的对比,选用特征价格法用于城市存量地下空间价值评估,以存量地下空间更新的微观形成机制估算潜在价值,采用定性分析与定量评价相结合的方式,在重要因素与更新利用之间建立回归模型,选择拟合度较好的半对数模型,并提出借助多元数据,提高评估的准确性,拟合存量地下空间的更新价值分布,以期为存量地下空间的高效、高品质更新利用提供参考。

城市街区单元绿地空间格局与植物群落碳汇效益优化研究

【目的】目前城市绿地已经进入存量发展阶段,如何在城市有限绿地空间条件下实现“碳中和”目标,合理布局低碳绿地空间格局,科学配置植物群落以增加生物多样性,成为目前城市绿地碳汇效益研究的重点和难点。【方法】以西安市碑林区和沣西新城的城市绿地空间格局为研究对象,对研究区域内街区单元的绿地空间格局进行分析并分类,探索街区单元绿地空间格局与碳汇量化关系,并从平面布局和垂直结构两方面提出城市街区单元绿地碳汇效益的优化方法,并选取碳储量最低的单核心辐射型街区单元的3个绿地样方进行碳汇效益优化设计。【结果】总结出4种西安市典型城市街区单元绿地空间格局模式,明晰了碳储量分布特征以及绿地空间格局对碳储量的影响机制,提出了城市街区单元绿地空间格局碳汇效益优化方法,发现碳储量与斑块类型面积(CA)、景观形状指数(LSI)呈极显著正相关关系,与聚集度(AI)呈极显著负相关关系。各类绿地空间格局碳储量大小为多核心辐射型>散点分布型或廊道穿越型>单核心辐射型,建议在单核心辐射型街区单元优化中增加植物群落层次结构,并增加高固碳植物种类及数量,提升样方内年固碳量。【结论】从中观尺度讨论街区单元绿地空间格局与植物群落碳汇效益之间的关系,使城市中小尺度绿地发挥应有的生态系统服务功能,探索街区单元绿地空间格局的构成要素与设计方法,为低碳绿地设计提供参考,提升城市街区单元内社区生活圈的人居环境质量。

基于PLUS和InVEST模型的合肥市生态系统碳储量时空演变特征

[目的]为寻求“双碳”目标导向下的合肥市城市发展新方案。[方法]依据合肥市2000-2020年5期土地利用数据,在合肥市国土空间规划指引下,运用PLUS模型模拟得到2035年合肥市土地利用时空演变规律,耦合InVEST模型探究多情景下合肥市碳储量时空变化特征,并进一步挖掘土地综合利用程度对碳储量的影响。[结果](1)2000-2020年合肥市土地利用变化特征主要表现为耕地、林地减少,其中耕地为建设用地扩增主要来源。自然发展和农田资源保护情景的土地变化规律大致相同,主要表现为耕地、林地、水体减少;绿色汇增城市发展情景下,林地相比其余2个情景面积由减少转为增加。(2)2000-2020年合肥市碳储量逐年递减,其中2005-2010年碳损失最为剧烈。到2035年,自然发展情景、农田资源保护情景、绿色汇增城市发展情景碳储量分别为138.96×10^(6),140.13×10^(6),139.81×10^(6) t。农田资源保护情景下,碳储量明显增加区域最低,建设用地扩张减缓;绿色汇增城市发展情景下,林地由碳损失转为碳固持,是最具固碳潜力的发展趋势。(3)绿色汇增城市发展情景土地利用率最高,可有效降低土地综合利用程度对碳储量流失的威胁。[结论]农田资源保护情景与绿色汇增城市发展情景均有助于城市固碳发展,实施生态保护与城市发展并行政策,调整土地综合利用模式有助于改善城市碳流失。

饲养密度对笼养白羽肉鸡生长性能、羽毛评分、血清生化和免疫指标的影响

试验旨在探究不同饲养密度对笼养白羽肉鸡生长性能、羽毛评分、血清生化和免疫指标的影响。试验选取504只体重相近、健康状况良好的1日龄科宝白羽肉鸡母雏,随机分为3组,每组8个重复(笼)。7~14日龄3组饲养密度分别为16只/m^(2)、17只/m^(2)、18只/m^(2),15~39日龄3组饲养密度为13只/m^(2)、14只/m^(2)、15只/m^(2)。结果显示:(1)不同饲养密度对7~14日龄和15~28日龄肉鸡平均采食量和料重比产生显著影响(P<0.05),且15~28日龄时,14只/m^(2)组平均日增重显著高于13只/m^(2)组(P<0.05),13只/m^(2)组料重比显著高于15只/m^(2)组(P<0.05)。29~39日龄13只/m^(2)组和14只/m^(2)组平均日增重显著高于15只/m^(2)(P<0.05);(2)不同饲养密度对肉鸡14日龄主翼羽长度产生显著影响(P<0.05);(3)不同饲养密度对肉鸡14日龄和39日龄羽毛评分产生显著影响(P<0.05);(4)不同饲养密度对肉鸡血清中碱性磷酸酶(ALP)活力、三碘甲状腺原氨酸(T3)和甲状腺素(T4)的含量产生显著影响(P<0.05),17只/m^(2)组显著高于16只/m^(2)组和18只/m^(2)组(P<0.05),14只/m^(2)组显著高于13只/m^(2)组和15只/m^(2)组(P<0.05);(5)不同饲养密度对肉鸡14日龄和39日龄血清中白细胞介素-4(IL-4)、白细胞介素-6(IL-6)、白细胞介素-10(IL-10)和白细胞介素-1β(IL-1β)含量产生显著影响(P<0.05)。研究表明,笼养科宝白羽肉鸡7~14日龄适宜饲养密度为17只/m^(2),15~39日龄适宜饲养密度为14只/m^(2)。

基于CBAMs-BiLSTM模型的中国股市预测

卷积块注意力模块(CBAM)因其可以有效地提高深度学习模型的预测准确性从而在各种预测问题中显示了其优越性。然而,CBAM在股指预测问题中的有效性研究却十分有限。为了解决这个问题并提高股指的预测精度,本文提出了CBAMs-BiLSTM模型。它将多个CBAM与双向长短期记忆网络(BiLSTM)相结合。研究中,标准指标评价法(SME)和模型置信集检验(MCS)用于综合评价模型的优越性和稳健性。实验数据为具有代表性的中国股指数据集:上证综合指数和深证综合指数。数值结果表明,CBAMs-BiLSTM优于单独的BiLSTM。其中在MAE,RMSE和MAPE上分别平均降低了13.06%,13.39%和12.48%。这证实了CBAM可以有效地提高BiLSTM的预测精度。此外,通过与其他流行模型进行对比,并研究了改变数据集、预测方法和训练集的大小的影响。结果一致证实了CBAMs-BiLSTM在预测精度和投资回报方面的优越性和稳健性。

VAE-ATTGRU模型的股指期货价格预测研究

针对股指期货市场高波动、非平稳、非线性和高信噪比等特性造成的预测难度大的问题,利用变分自编码器(VAE)和循环神经网络(RNN)提出一种基于VAE-ATTGRU的混合深度学习股指期货价格预测模型。利用变分自编码器对股指期货技术指标进行学习,将VAE学习到的潜在因子与原始数据融合实现数据增强,得到更丰富的因子表示;使用循环神经网络对股指期货价格进行预测,发现结合了注意力机制的门控循环单元(ATTGRU)可以对VAE增强后的股指期货数据进行充分学习,对关键特征信息进行捕捉并重新赋予权重。在沪深300股指期货、中证500股指期货和上证50股指期货数据上进行实验,通过均方根误差(RMSE)、平均绝对误差(MAE)、平均绝对百分比误差(MAPE)和决定系数R2对VAE-ATTGRU模型进行评估,发现其在预测精度上优于其他模型。

基于MVC5B混合模型的中国股指预测研究

为了提高中国股指的预测表现,提出一个融合了变分模态分解(VMD)、卷积注意力模块(CBAM)和双向长短期记忆网络(BiLSTM)的混合模型MVC5B(multi-channel-VMD-CBAM5-BiLSTM)。不同于混合模型常用的分解-集成构造方法,MVC5B基于提出的多通道输入方法构造而成。多通道输入方法基于自身一次性预测的特点可以有效规避分解-集成方法多次预测带来的累计误差和巨大计算成本,从而提升MVC5B的预测表现。CBAM的引入不但提升了股指的预测表现,而且还丰富了股指预测问题中关于CBAM的研究。基于多个具有代表性的中国股指数据集的实证结果显示MVC5B的预测表现和模拟收益显著优于流行的预测模型。实证结果还进一步证实了多通道输入方法相比于分解-集成方法的优越性以及CBAM在股指预测问题中的有效性。

基于无人机影像的高郁闭度马尾松林蓄积量估算方法研究

为降低传统野外调查成本,提高高郁闭度森林资源调查效率,以多光谱无人机影像结合实地样地调查数据为源数据,以马尾松纯林为研究对象,采用冠层高度模型(CHM)及多光谱影像6种植被指数构建模型,估算研究区林分蓄积量。结果表明:1)高分辨率数字高程模型的辅助能够有效弥补无人机影像在茂密森林无法提取地面点的缺陷,可提高CHM构建的精度,实现在茂密森林树高的准确提取;2)采用CHM提取研究区单木树高并估算蓄积量时,样地内共提取马尾松292株,提取的平均树高为18.77 m,小班区域内共提取马尾松18 120株,提取的平均树高为17.02 m,实测平均树高为18.17 m,平均树高提取效果较好,估算蓄积量为7 466.74 m3,实测蓄积量为9 024.40 m3,蓄积量估算精度为82.90%;3)植被指数模型的RMSE为0.39,R2为0.84,模型精度较高,蓄积量估算为8 620.30 m3,估算精度为96.26%。通过借助无人机遥感技术,两种蓄积量估算方法均能在高郁闭度森林中实现蓄积量的快速估算,其中通过多光谱数据提取植被指数构建模型估算蓄积量的效果更佳,这为进一步推广无人机影像在高郁闭度森林资源调查中的应用提供了有力的科学依据。

金融科技对股指期货流动性的影响

本文研究了金融科技对沪深300股指期货流动性的影响,基于2017—2022年沪深300股指期货的交易数据,通过分位数回归方法分析并得到以下结论:金融科技能够显著影响股指期货市场的流动性水平;金融科技对期货市场流动性的影响是动态变化的,并且可能由于效果和时间层面的不同而产生变化。

基于深度学习的股票价格指数预测

文章以2012—2022年上证50、沪深300、中证500三支股指交易数据为研究对象,利用深度学习预测股指,分别构建了卷积神经网络(CNN)模型、长短期记忆神经网络(LSTM)模型和CNN-LSTM组合神经网络模型,使用均方根误差(RMSE)、平均绝对误差(MAE)、平均百分比误差绝对值(MAPE)和决定系数(R2)4个指标对模型进行评价。结果表明,组合模型兼有CNN模型的高维特征挖掘优势和LSTM模型的时序特征提取优势,在3支股指的预测中,组合模型的预测精度均高于CNN模型和LSTM模型单一模型,其RMSE指标分别提升了28.60百分点、52.56百分点和25.28百分点,证实了CNN-LSTM组合神经网络模型的有效性和准确性。

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。

新质生产力股票指数的设计与应用

本文通过综合考量研发投入、生产效率、环境影响及社会责任等多个维度的指标,选取了在A股、港股上市的50家具备显著新质生产力特征的上市公司,并据此构建了一个活跃交易、规模可观且适合投资的新质生产力股票指数。该指数旨在通过评估企业的新质生产力水平来衡量这些企业长期盈利能力与股价表现。在权重分配方面,本研究结合了主成分分析法、熵值法和市值法,不仅考量了公司的现有市值,更侧重于由新质生产力带来的长期增长潜力的积累。对历史数据进行回溯测试的结果显示,基于新质生产力特征构建的股票指数总体表现优于传统市值加权指数,展现出更高的收益率和更低的波动性。这一发现凸显了新质生产力对于企业高质量发展的关键作用。尤其是在快速变化的市场环境中,企业若能以提升自身的新质生产力为基石,通过技术创新和资源优化配置,将有望增强市场竞争力并获得更高的财务回报。因此,投资者可以将新质生产力作为选股的重要参考,通过投资拥有高水平新质生产力的企业,获得更佳的投资回报。