首页 > 最新文献

arXiv - QuantFin - Statistical Finance最新文献

英文 中文
Trading Devil Final: Backdoor attack via Stock market and Bayesian Optimization 交易魔鬼决赛:通过股市和贝叶斯优化进行后门攻击
Pub Date : 2024-07-21 DOI: arxiv-2407.14573
Orson Mengara
Since the advent of generative artificial intelligence, every company andresearcher has been rushing to develop their own generative models, whethercommercial or not. Given the large number of users of these powerful new tools,there is currently no intrinsically verifiable way to explain from the groundup what happens when LLMs (large language models) learn. For example, thosebased on automatic speech recognition systems, which have to rely on huge andastronomical amounts of data collected from all over the web to produce fastand efficient results, In this article, we develop a backdoor attack calledMarketBackFinal 2.0, based on acoustic data poisoning, MarketBackFinal 2.0 ismainly based on modern stock market models. In order to show the possiblevulnerabilities of speech-based transformers that may rely on LLMs.
自从生成式人工智能问世以来,每家公司和研究人员都急于开发自己的生成式模型,无论是否具有商业价值。鉴于这些功能强大的新工具有大量用户,目前还没有内在可验证的方法来从根本上解释 LLM(大型语言模型)学习时会发生什么。例如,那些基于自动语音识别的系统,它们必须依赖从网络上收集的大量数据才能快速高效地生成结果,而在本文中,我们开发了一种名为 MarketBackFinal 2.0 的基于声学数据中毒的后门攻击,MarketBackFinal 2.0 主要基于现代股票市场模型。为了展示基于语音的转换器可能存在的漏洞,这些转换器可能依赖于 LLMs。
{"title":"Trading Devil Final: Backdoor attack via Stock market and Bayesian Optimization","authors":"Orson Mengara","doi":"arxiv-2407.14573","DOIUrl":"https://doi.org/arxiv-2407.14573","url":null,"abstract":"Since the advent of generative artificial intelligence, every company and\u0000researcher has been rushing to develop their own generative models, whether\u0000commercial or not. Given the large number of users of these powerful new tools,\u0000there is currently no intrinsically verifiable way to explain from the ground\u0000up what happens when LLMs (large language models) learn. For example, those\u0000based on automatic speech recognition systems, which have to rely on huge and\u0000astronomical amounts of data collected from all over the web to produce fast\u0000and efficient results, In this article, we develop a backdoor attack called\u0000MarketBackFinal 2.0, based on acoustic data poisoning, MarketBackFinal 2.0 is\u0000mainly based on modern stock market models. In order to show the possible\u0000vulnerabilities of speech-based transformers that may rely on LLMs.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"16 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141774364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Super-efficiency and Stock Market Valuation: Evidence from Listed Banks in China (2006 to 2023) 超效率与股市估值:中国上市银行的证据(2006 年至 2023 年)
Pub Date : 2024-07-20 DOI: arxiv-2407.14734
Yun Liao
This study investigates the relationship between bank efficiency and stockmarket valuation using an unbalanced panel dataset of 42 listed banks in Chinafrom 2006 to 2023. We employ a non-radial and non-oriented slack basedsuper-efficiency Data Envelopment Analysis (Super-SBM-UND-VRS based DEA) model,which treats Non-Performing Loans (NPLs) as an undesired output. Our resultsshow that the relationship between super-efficiency and stock market valuationis stronger than that between Return on Asset (ROA) and stock marketperformance, as measured by Tobin's Q. Notably, the Super-SBM-UND-VRS modelyields novel results compared to other efficiency methods, such as theStochastic Frontier Analysis (SFA) approach and traditional DEA models.Furthermore, our results suggest that bank evaluations benefit from decreasedownership concentration, whereas interest rate liberalization has the oppositeeffect.
本研究使用 2006 年至 2023 年中国 42 家上市银行的非平衡面板数据集研究银行效率与股票市场估值之间的关系。我们采用了基于非径向和非定向松弛的超效率数据包络分析(Super-SBM-UND-VRS based DEA)模型,该模型将不良贷款(NPLs)视为非期望产出。我们的结果表明,超效率与股票市场估值之间的关系强于资产收益率(ROA)与托宾 Q 衡量的股票市场表现之间的关系。值得注意的是,与其他效率方法(如托氏前沿分析法(SFA)和传统 DEA 模型)相比,超 SBM-UND-VRS 模式产生了新颖的结果。
{"title":"Super-efficiency and Stock Market Valuation: Evidence from Listed Banks in China (2006 to 2023)","authors":"Yun Liao","doi":"arxiv-2407.14734","DOIUrl":"https://doi.org/arxiv-2407.14734","url":null,"abstract":"This study investigates the relationship between bank efficiency and stock\u0000market valuation using an unbalanced panel dataset of 42 listed banks in China\u0000from 2006 to 2023. We employ a non-radial and non-oriented slack based\u0000super-efficiency Data Envelopment Analysis (Super-SBM-UND-VRS based DEA) model,\u0000which treats Non-Performing Loans (NPLs) as an undesired output. Our results\u0000show that the relationship between super-efficiency and stock market valuation\u0000is stronger than that between Return on Asset (ROA) and stock market\u0000performance, as measured by Tobin's Q. Notably, the Super-SBM-UND-VRS model\u0000yields novel results compared to other efficiency methods, such as the\u0000Stochastic Frontier Analysis (SFA) approach and traditional DEA models.\u0000Furthermore, our results suggest that bank evaluations benefit from decreased\u0000ownership concentration, whereas interest rate liberalization has the opposite\u0000effect.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"64 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141774245","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Sentiment Analysis of State Bank of Pakistan's Monetary Policy Documents and its Impact on Stock Market 巴基斯坦国家银行货币政策文件的情绪分析及其对股市的影响
Pub Date : 2024-07-19 DOI: arxiv-2408.03328
Aabid Karim, Heman Das Lohano
This research examines whether sentiments conveyed in the State Bank ofPakistan's (SBP) communications impact financial market expectations and canact as a monetary policy tool. To achieve our goal, we first use sentimentanalysis techniques to quantify the tone of SBP monetary policy documents andsecond, we use short time window, high frequency methodology to approximate theimpact of tone on stock market returns. Our results show that positive(negative) change in the tone positively (negatively) impacts stock returns inKarachi Stock Exchange. Further extension shows that the communication of SBPstill has a statistically significant impact on stock returns when controllingfor different variables and monetary policy tool. Also, the communication ofSBP does not have a long term constant effect on stock market.
本研究探讨了巴基斯坦国家银行(SBP)通信中所传达的情绪是否会影响金融市场预期,以及能否作为一种货币政策工具发挥作用。为了实现我们的目标,我们首先使用情绪分析技术来量化巴基斯坦国家银行货币政策文件的基调;其次,我们使用短时间窗口、高频率方法来近似分析基调对股市回报的影响。我们的研究结果表明,基调的积极(消极)变化会对卡拉奇证券交易所的股票回报率产生积极(消极)影响。进一步的扩展表明,在控制不同变量和货币政策工具的情况下,印度央行的沟通仍然对股票回报率有显著的统计影响。此外,印度央行的沟通对股票市场也没有长期不变的影响。
{"title":"Sentiment Analysis of State Bank of Pakistan's Monetary Policy Documents and its Impact on Stock Market","authors":"Aabid Karim, Heman Das Lohano","doi":"arxiv-2408.03328","DOIUrl":"https://doi.org/arxiv-2408.03328","url":null,"abstract":"This research examines whether sentiments conveyed in the State Bank of\u0000Pakistan's (SBP) communications impact financial market expectations and can\u0000act as a monetary policy tool. To achieve our goal, we first use sentiment\u0000analysis techniques to quantify the tone of SBP monetary policy documents and\u0000second, we use short time window, high frequency methodology to approximate the\u0000impact of tone on stock market returns. Our results show that positive\u0000(negative) change in the tone positively (negatively) impacts stock returns in\u0000Karachi Stock Exchange. Further extension shows that the communication of SBP\u0000still has a statistically significant impact on stock returns when controlling\u0000for different variables and monetary policy tool. Also, the communication of\u0000SBP does not have a long term constant effect on stock market.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141933531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Risk Analysis of Passive Portfolios 被动投资组合的风险分析
Pub Date : 2024-07-11 DOI: arxiv-2407.08332
Sourish Das
In this work, we present an alternative passive investment strategy. Thepassive investment philosophy comes from the Efficient Market Hypothesis (EMH),and its adoption is widespread. If EMH is true, one cannot outperform market byactively managing their portfolio for a long time. Also, it requires little tono intervention. People can buy an exchange-traded fund (ETF) with a long-termperspective. As the economy grows over time, one expects the ETF to grow. Forexample, in India, one can invest in NETF, which suppose to mimic the Nifty50return. However, the weights of the Nifty 50 index are based on marketcapitalisation. These weights are not necessarily optimal for the investor. Inthis work, we present that volatility risk and extreme risk measures of theNifty50 portfolio are uniformly larger than Markowitz's optimal portfolio.However, common people can't create an optimised portfolio. So we proposed analternative passive investment strategy of an equal-weight portfolio. We showthat if one pushes the maximum weight of the portfolio towards equal weight,the idiosyncratic risk of the portfolio would be minimal. The empiricalevidence indicates that the risk profile of an equal-weight portfolio issimilar to that of Markowitz's optimal portfolio. Hence instead of buyingNifty50 ETFs, one should equally invest in the stocks of Nifty50 to achieve auniformly better risk profile than the Nifty 50 ETF portfolio. We also presentan analysis of how portfolios perform to idiosyncratic events like the Russianinvasion of Ukraine. We found that the equal weight portfolio has a uniformlylower risk than the Nifty 50 portfolio before and during the Russia-Ukrainewar. All codes are available on GitHub(url{https://github.com/sourish-cmi/quant/tree/main/Chap_Risk_Anal_of_Passive_Portfolio}).
在这项工作中,我们提出了另一种被动投资策略。被动投资理念源于有效市场假说(Efficient Market Hypothesis,EMH),并被广泛采用。如果 EMH 属实,那么人们就无法通过长期主动管理投资组合来跑赢市场。而且,它几乎不需要干预。人们可以购买具有长期观点的交易所交易基金(ETF)。随着时间的推移,经济增长,人们预期 ETF 也会增长。例如,在印度,人们可以投资于NETF,该基金假定模仿Nifty50的回报率。但是,Nifty 50 指数的权重是基于市值的。对于投资者来说,这些权重并不一定是最优的。在这项研究中,我们发现 Nifty50 指数投资组合的波动风险和极端风险度量均大于马科维茨的最优投资组合。因此,我们提出了等权重投资组合的替代性被动投资策略。我们的研究表明,如果将投资组合的最大权重推向等权重,投资组合的特异性风险将降到最低。经验证据表明,等权重投资组合的风险状况与马科维茨的最优投资组合相似。因此,与其购买 Nifty50 ETF,不如平均投资 Nifty50 的股票,以获得比 Nifty 50 ETF 投资组合更好的风险状况。我们还分析了投资组合在俄罗斯入侵乌克兰等特殊事件中的表现。我们发现,在俄乌战争之前和期间,等权重投资组合的风险均低于 Nifty 50 投资组合。所有代码均可在 GitHub 上获取(url{https://github.com/sourish-cmi/quant/tree/main/Chap_Risk_Anal_of_Passive_Portfolio})。
{"title":"Risk Analysis of Passive Portfolios","authors":"Sourish Das","doi":"arxiv-2407.08332","DOIUrl":"https://doi.org/arxiv-2407.08332","url":null,"abstract":"In this work, we present an alternative passive investment strategy. The\u0000passive investment philosophy comes from the Efficient Market Hypothesis (EMH),\u0000and its adoption is widespread. If EMH is true, one cannot outperform market by\u0000actively managing their portfolio for a long time. Also, it requires little to\u0000no intervention. People can buy an exchange-traded fund (ETF) with a long-term\u0000perspective. As the economy grows over time, one expects the ETF to grow. For\u0000example, in India, one can invest in NETF, which suppose to mimic the Nifty50\u0000return. However, the weights of the Nifty 50 index are based on market\u0000capitalisation. These weights are not necessarily optimal for the investor. In\u0000this work, we present that volatility risk and extreme risk measures of the\u0000Nifty50 portfolio are uniformly larger than Markowitz's optimal portfolio.\u0000However, common people can't create an optimised portfolio. So we proposed an\u0000alternative passive investment strategy of an equal-weight portfolio. We show\u0000that if one pushes the maximum weight of the portfolio towards equal weight,\u0000the idiosyncratic risk of the portfolio would be minimal. The empirical\u0000evidence indicates that the risk profile of an equal-weight portfolio is\u0000similar to that of Markowitz's optimal portfolio. Hence instead of buying\u0000Nifty50 ETFs, one should equally invest in the stocks of Nifty50 to achieve a\u0000uniformly better risk profile than the Nifty 50 ETF portfolio. We also present\u0000an analysis of how portfolios perform to idiosyncratic events like the Russian\u0000invasion of Ukraine. We found that the equal weight portfolio has a uniformly\u0000lower risk than the Nifty 50 portfolio before and during the Russia-Ukraine\u0000war. All codes are available on GitHub\u0000(url{https://github.com/sourish-cmi/quant/tree/main/Chap_Risk_Anal_of_Passive_Portfolio}).","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"18 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141612154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Multiple split approach -- multidimensional probabilistic forecasting of electricity markets 多重分割法--电力市场的多维概率预测
Pub Date : 2024-07-10 DOI: arxiv-2407.07795
Katarzyna Maciejowska, Weronika Nitka
In this article, a multiple split method is proposed that enablesconstruction of multidimensional probabilistic forecasts of a selected set ofvariables. The method uses repeated resampling to estimate uncertainty ofsimultaneous multivariate predictions. This nonparametric approach links thegap between point and probabilistic predictions and can be combined withdifferent point forecasting methods. The performance of the method is evaluatedwith data describing the German short-term electricity market. The results showthat the proposed approach provides highly accurate predictions. The gains frommultidimensional forecasting are the largest when functions of variables, suchas price spread or residual load, are considered. Finally, the method is used to support a decision process of a moderategeneration utility that produces electricity from wind energy and sells it oneither a day-ahead or an intraday market. The company makes decisions underhigh uncertainty because it knows neither the future production level nor theprices. We show that joint forecasting of both market prices and fundamentalscan be used to predict the distribution of a profit, and hence helps to designa strategy that balances a level of income and a trading risk.
本文提出了一种多重分割方法,该方法能够构建对一组选定变量的多维概率预测。该方法使用重复重采样来估计同时多变量预测的不确定性。这种非参数方法将点预测与概率预测之间的差距联系起来,并可与不同的点预测方法相结合。我们利用德国短期电力市场的数据对该方法的性能进行了评估。结果表明,所提出的方法能提供高度准确的预测。在考虑价格差或剩余负荷等变量函数时,多维预测的收益最大。最后,该方法被用于支持一家利用风能发电并在日前或当日市场上销售的中型发电公司的决策过程。该公司在高度不确定的情况下做出决策,因为它既不知道未来的生产水平,也不知道价格。我们的研究表明,对市场价格和基本面的联合预测可以用来预测利润的分配,从而有助于设计一种平衡收益水平和交易风险的策略。
{"title":"Multiple split approach -- multidimensional probabilistic forecasting of electricity markets","authors":"Katarzyna Maciejowska, Weronika Nitka","doi":"arxiv-2407.07795","DOIUrl":"https://doi.org/arxiv-2407.07795","url":null,"abstract":"In this article, a multiple split method is proposed that enables\u0000construction of multidimensional probabilistic forecasts of a selected set of\u0000variables. The method uses repeated resampling to estimate uncertainty of\u0000simultaneous multivariate predictions. This nonparametric approach links the\u0000gap between point and probabilistic predictions and can be combined with\u0000different point forecasting methods. The performance of the method is evaluated\u0000with data describing the German short-term electricity market. The results show\u0000that the proposed approach provides highly accurate predictions. The gains from\u0000multidimensional forecasting are the largest when functions of variables, such\u0000as price spread or residual load, are considered. Finally, the method is used to support a decision process of a moderate\u0000generation utility that produces electricity from wind energy and sells it on\u0000either a day-ahead or an intraday market. The company makes decisions under\u0000high uncertainty because it knows neither the future production level nor the\u0000prices. We show that joint forecasting of both market prices and fundamentals\u0000can be used to predict the distribution of a profit, and hence helps to design\u0000a strategy that balances a level of income and a trading risk.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"31 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141586463","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Advanced Financial Fraud Detection Using GNN-CL Model 利用 GNN-CL 模型进行高级金融欺诈检测
Pub Date : 2024-07-09 DOI: arxiv-2407.06529
Yu Cheng, Junjie Guo, Shiqing Long, You Wu, Mengfang Sun, Rong Zhang
The innovative GNN-CL model proposed in this paper marks a breakthrough inthe field of financial fraud detection by synergistically combining theadvantages of graph neural networks (gnn), convolutional neural networks (cnn)and long short-term memory (LSTM) networks. This convergence enablesmultifaceted analysis of complex transaction patterns, improving detectionaccuracy and resilience against complex fraudulent activities. A key novelty ofthis paper is the use of multilayer perceptrons (MLPS) to estimate nodesimilarity, effectively filtering out neighborhood noise that can lead to falsepositives. This intelligent purification mechanism ensures that only the mostrelevant information is considered, thereby improving the model's understandingof the network structure. Feature weakening often plagues graph-based modelsdue to the dilution of key signals. In order to further address the challengeof feature weakening, GNN-CL adopts reinforcement learning strategies. Bydynamically adjusting the weights assigned to central nodes, it reinforces theimportance of these influential entities to retain important clues of fraudeven in less informative data. Experimental evaluations on Yelp datasets showthat the results highlight the superior performance of GNN-CL compared toexisting methods.
本文提出的 GNN-CL 创新模型将图神经网络(gnn)、卷积神经网络(cnn)和长短期记忆(LSTM)网络的优势协同结合在一起,标志着金融欺诈检测领域的一项突破。这种融合能够对复杂的交易模式进行多方面分析,从而提高检测精度和抵御复杂欺诈活动的能力。本文的一个主要创新点是使用多层感知器(MLPS)来估计节点相似性,从而有效过滤掉可能导致误判的邻域噪声。这种智能净化机制确保只考虑最相关的信息,从而提高了模型对网络结构的理解。由于关键信号被稀释,基于图的模型经常受到特征弱化的困扰。为了进一步解决特征弱化的难题,GNN-CL 采用了强化学习策略。通过动态调整分配给中心节点的权重,它强化了这些有影响力实体的重要性,即使在信息量较少的数据中也能保留重要的欺诈线索。在 Yelp 数据集上的实验评估结果表明,与现有方法相比,GNN-CL 的性能更优越。
{"title":"Advanced Financial Fraud Detection Using GNN-CL Model","authors":"Yu Cheng, Junjie Guo, Shiqing Long, You Wu, Mengfang Sun, Rong Zhang","doi":"arxiv-2407.06529","DOIUrl":"https://doi.org/arxiv-2407.06529","url":null,"abstract":"The innovative GNN-CL model proposed in this paper marks a breakthrough in\u0000the field of financial fraud detection by synergistically combining the\u0000advantages of graph neural networks (gnn), convolutional neural networks (cnn)\u0000and long short-term memory (LSTM) networks. This convergence enables\u0000multifaceted analysis of complex transaction patterns, improving detection\u0000accuracy and resilience against complex fraudulent activities. A key novelty of\u0000this paper is the use of multilayer perceptrons (MLPS) to estimate node\u0000similarity, effectively filtering out neighborhood noise that can lead to false\u0000positives. This intelligent purification mechanism ensures that only the most\u0000relevant information is considered, thereby improving the model's understanding\u0000of the network structure. Feature weakening often plagues graph-based models\u0000due to the dilution of key signals. In order to further address the challenge\u0000of feature weakening, GNN-CL adopts reinforcement learning strategies. By\u0000dynamically adjusting the weights assigned to central nodes, it reinforces the\u0000importance of these influential entities to retain important clues of fraud\u0000even in less informative data. Experimental evaluations on Yelp datasets show\u0000that the results highlight the superior performance of GNN-CL compared to\u0000existing methods.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141576491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Stochastic Approaches to Asset Price Analysis 资产价格分析的随机方法
Pub Date : 2024-07-09 DOI: arxiv-2407.06745
Michael Sekatchev, Zhengxiang Zhou
In this project, we propose to explore the Kalman filter's performance forestimating asset prices. We begin by introducing a stochastic mean-revertingprocesses, the Ornstein-Uhlenbeck (OU) model. After this we discuss the Kalmanfilter in detail, and its application with this model. After a demonstration ofthe Kalman filter on a simulated OU process and a discussion of maximumlikelihood estimation (MLE) for estimating model parameters, we apply theKalman filter with the OU process and trailing parameter estimation to realstock market data. We finish by proposing a simple day-trading algorithm usingthe Kalman filter with the OU process and backtest its performance usingApple's stock price. We then move to the Heston model, a combination ofGeometric Brownian Motion and the OU process. Maximum likelihood estimation iscommonly used for Heston model parameter estimation, which results in verycomplex forms. Here we propose an alternative but easier way of parameterestimation, called the method of moments (MOM). After the derivation of theseestimators, we again apply this method to real stock data to assess itsperformance.
在本项目中,我们建议探索卡尔曼滤波器在估计资产价格方面的性能。我们首先介绍一个随机均值回复过程,即 Ornstein-Uhlenbeck (OU) 模型。之后,我们将详细讨论卡尔曼滤波器及其在该模型中的应用。在演示了卡尔曼滤波器在模拟 OU 过程中的应用,并讨论了用于估计模型参数的最大似然估计 (MLE)之后,我们将卡尔曼滤波器与 OU 过程和跟踪参数估计一起应用于真实股市数据。最后,我们提出了一种使用卡尔曼滤波和 OU 过程的简单日内交易算法,并使用苹果公司的股票价格对其性能进行了回溯测试。然后,我们转向赫斯顿模型,这是几何布朗运动和 OU 过程的结合。最大似然估计法通常用于赫斯顿模型参数估计,这会导致非常复杂的形式。在此,我们提出了另一种更简便的参数估计方法,即矩法(MOM)。在推导出这些估计方法后,我们再次将该方法应用于真实股票数据,以评估其性能。
{"title":"Stochastic Approaches to Asset Price Analysis","authors":"Michael Sekatchev, Zhengxiang Zhou","doi":"arxiv-2407.06745","DOIUrl":"https://doi.org/arxiv-2407.06745","url":null,"abstract":"In this project, we propose to explore the Kalman filter's performance for\u0000estimating asset prices. We begin by introducing a stochastic mean-reverting\u0000processes, the Ornstein-Uhlenbeck (OU) model. After this we discuss the Kalman\u0000filter in detail, and its application with this model. After a demonstration of\u0000the Kalman filter on a simulated OU process and a discussion of maximum\u0000likelihood estimation (MLE) for estimating model parameters, we apply the\u0000Kalman filter with the OU process and trailing parameter estimation to real\u0000stock market data. We finish by proposing a simple day-trading algorithm using\u0000the Kalman filter with the OU process and backtest its performance using\u0000Apple's stock price. We then move to the Heston model, a combination of\u0000Geometric Brownian Motion and the OU process. Maximum likelihood estimation is\u0000commonly used for Heston model parameter estimation, which results in very\u0000complex forms. Here we propose an alternative but easier way of parameter\u0000estimation, called the method of moments (MOM). After the derivation of these\u0000estimators, we again apply this method to real stock data to assess its\u0000performance.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141576489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
International Trade Flow Prediction with Bilateral Trade Provisions 利用双边贸易条款预测国际贸易流量
Pub Date : 2024-06-23 DOI: arxiv-2407.13698
Zijie Pan, Stepan Gordeev, Jiahui Zhao, Ziyi Meng, Caiwen Ding, Sandro Steinbach, Dongjin Song
This paper presents a novel methodology for predicting internationalbilateral trade flows, emphasizing the growing importance of Preferential TradeAgreements (PTAs) in the global trade landscape. Acknowledging the limitationsof traditional models like the Gravity Model of Trade, this study introduces atwo-stage approach combining explainable machine learning and factorizationmodels. The first stage employs SHAP Explainer for effective variableselection, identifying key provisions in PTAs, while the second stage utilizesFactorization Machine models to analyze the pairwise interaction effects ofthese provisions on trade flows. By analyzing comprehensive datasets, the paperdemonstrates the efficacy of this approach. The findings not only enhance thepredictive accuracy of trade flow models but also offer deeper insights intothe complex dynamics of international trade, influenced by specific bilateraltrade provisions.
本文提出了一种预测国际双边贸易流量的新方法,强调了优惠贸易协定(PTAs)在全球贸易格局中日益增长的重要性。考虑到传统模型(如贸易引力模型)的局限性,本研究引入了结合可解释机器学习和因式分解模型的两阶段方法。第一阶段采用 SHAP Explainer 进行有效的变量选择,识别出 PTA 中的关键条款;第二阶段采用因子化机器模型分析这些条款对贸易流量的成对交互效应。通过分析综合数据集,本文证明了这种方法的有效性。研究结果不仅提高了贸易流量模型的预测准确性,还深入揭示了国际贸易受特定双边贸易条款影响的复杂动态。
{"title":"International Trade Flow Prediction with Bilateral Trade Provisions","authors":"Zijie Pan, Stepan Gordeev, Jiahui Zhao, Ziyi Meng, Caiwen Ding, Sandro Steinbach, Dongjin Song","doi":"arxiv-2407.13698","DOIUrl":"https://doi.org/arxiv-2407.13698","url":null,"abstract":"This paper presents a novel methodology for predicting international\u0000bilateral trade flows, emphasizing the growing importance of Preferential Trade\u0000Agreements (PTAs) in the global trade landscape. Acknowledging the limitations\u0000of traditional models like the Gravity Model of Trade, this study introduces a\u0000two-stage approach combining explainable machine learning and factorization\u0000models. The first stage employs SHAP Explainer for effective variable\u0000selection, identifying key provisions in PTAs, while the second stage utilizes\u0000Factorization Machine models to analyze the pairwise interaction effects of\u0000these provisions on trade flows. By analyzing comprehensive datasets, the paper\u0000demonstrates the efficacy of this approach. The findings not only enhance the\u0000predictive accuracy of trade flow models but also offer deeper insights into\u0000the complex dynamics of international trade, influenced by specific bilateral\u0000trade provisions.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"55 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141740145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Role of fee choice in revenue generation of AMMs: A quantitative study 收费选择在 AMM 创收中的作用:定量研究
Pub Date : 2024-06-18 DOI: arxiv-2406.12417
Abe Alexander, Jesse Moestaredjo, Mart Heuvelmans, Lars Fritz
In the ever evolving landscape of decentralized finance automated marketmakers (AMMs) play a key role: they provide a market place for trading assetsin a decentralized manner. For so-called bluechip pairs, arbitrage activityprovides a major part of the revenue generation of AMMs but also a major sourceof loss due to the so-called 'informed orderflow'. Finding ways to minimizethose losses while still keeping uninformed trading activity alive is a majorproblem in the field. In this paper we will investigate the mechanics of saidarbitrage and try to understand how AMMs can maximize the revenue creation orin other words minimize the losses. To that end, we model the dynamics ofarbitrage activity for a concrete implementation of a pool and study itssensitivity to the choice of fee aiming to maximize the revenue for the AMM. Weidentify dynamical fees that mimic the directionality of the price due toasymmetric fee choices as a promising avenue to mitigate losses to toxic flow.This work is based on and extends a recent article by some of the authors.
在不断发展的去中心化金融领域,自动做市商(AMM)扮演着重要角色:它们以去中心化的方式为资产交易提供了一个市场。对于所谓的蓝筹股对来说,套利活动是自动做市商创收的主要部分,但也是所谓的 "知情订单流 "造成损失的主要来源。如何在保持无信息交易活动的同时最大限度地减少这些损失,是该领域的一大难题。在本文中,我们将研究上述套利的机制,并试图了解 AMM 如何最大限度地创造收益,或者换句话说,如何最大限度地减少损失。为此,我们建立了一个具体实施池的套利活动动态模型,并研究了它对旨在使 AMM 收益最大化的费用选择的敏感性。我们认为,动态收费可以模仿不对称收费选择导致的价格方向性,是减轻毒流损失的一条可行途径。
{"title":"Role of fee choice in revenue generation of AMMs: A quantitative study","authors":"Abe Alexander, Jesse Moestaredjo, Mart Heuvelmans, Lars Fritz","doi":"arxiv-2406.12417","DOIUrl":"https://doi.org/arxiv-2406.12417","url":null,"abstract":"In the ever evolving landscape of decentralized finance automated market\u0000makers (AMMs) play a key role: they provide a market place for trading assets\u0000in a decentralized manner. For so-called bluechip pairs, arbitrage activity\u0000provides a major part of the revenue generation of AMMs but also a major source\u0000of loss due to the so-called 'informed orderflow'. Finding ways to minimize\u0000those losses while still keeping uninformed trading activity alive is a major\u0000problem in the field. In this paper we will investigate the mechanics of said\u0000arbitrage and try to understand how AMMs can maximize the revenue creation or\u0000in other words minimize the losses. To that end, we model the dynamics of\u0000arbitrage activity for a concrete implementation of a pool and study its\u0000sensitivity to the choice of fee aiming to maximize the revenue for the AMM. We\u0000identify dynamical fees that mimic the directionality of the price due to\u0000asymmetric fee choices as a promising avenue to mitigate losses to toxic flow.\u0000This work is based on and extends a recent article by some of the authors.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"53 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141503578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Modeling a Financial System with Memory via Fractional Calculus and Fractional Brownian Motion 通过分数微积分和分数布朗运动模拟有记忆的金融系统
Pub Date : 2024-06-12 DOI: arxiv-2406.19408
Patrick Geraghty
Financial markets have long since been modeled using stochastic methods suchas Brownian motion, and more recently, rough volatility models have been builtusing fractional Brownian motion. This fractional aspect brings memory into thesystem. In this project, we describe and analyze a financial model based on thefractional Langevin equation with colored noise generated by fractionalBrownian motion. Physics-based methods of analysis are used to examine thephase behavior and dispersion relations of the system upon varying inputparameters. A type of anomalous marginal glass phase is potentially seen insome regions, which motivates further exploration of this model and expandeduse of phase behavior and dispersion relation methods to analyze financialmodels.
长期以来,金融市场一直使用随机方法(如布朗运动)来建模,最近,人们又利用分 子布朗运动建立了粗略波动模型。这种分数运动为系统带来了记忆。在本项目中,我们描述并分析了一个基于分式朗文方程的金融模型,该模型带有由分式布朗运动产生的彩色噪声。我们采用了基于物理学的分析方法来研究输入参数变化时系统的相位行为和分散关系。在某些区域可能会出现一种反常的边缘玻璃相,这促使我们进一步探索这一模型,并扩大使用相行为和分散关系方法来分析金融模型。
{"title":"Modeling a Financial System with Memory via Fractional Calculus and Fractional Brownian Motion","authors":"Patrick Geraghty","doi":"arxiv-2406.19408","DOIUrl":"https://doi.org/arxiv-2406.19408","url":null,"abstract":"Financial markets have long since been modeled using stochastic methods such\u0000as Brownian motion, and more recently, rough volatility models have been built\u0000using fractional Brownian motion. This fractional aspect brings memory into the\u0000system. In this project, we describe and analyze a financial model based on the\u0000fractional Langevin equation with colored noise generated by fractional\u0000Brownian motion. Physics-based methods of analysis are used to examine the\u0000phase behavior and dispersion relations of the system upon varying input\u0000parameters. A type of anomalous marginal glass phase is potentially seen in\u0000some regions, which motivates further exploration of this model and expanded\u0000use of phase behavior and dispersion relation methods to analyze financial\u0000models.","PeriodicalId":501139,"journal":{"name":"arXiv - QuantFin - Statistical Finance","volume":"80 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141503474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
arXiv - QuantFin - Statistical Finance
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1