arXiv - QuantFin - Risk Management最新文献_第6页

An elementary proof of representation of submodular function as an supremum of measures on $σ$-algebra with totally ordered generating class 子模函数表示为具有完全有序生成类的 $σ$-algebra 上的量的上集的基本证明

arXiv - QuantFin - Risk Management Pub Date : 2024-06-26 DOI: arxiv-2406.18174

Tetsuya Hattori

引用次数: 0

Catastrophic-risk-aware reinforcement learning with extreme-value-theory-based policy gradients 基于极值理论策略梯度的灾难风险意识强化学习

arXiv - QuantFin - Risk Management Pub Date : 2024-06-21 DOI: arxiv-2406.15612

Parisa Davar, Frédéric Godin, Jose Garrido

引用次数: 0

Lessons From Model Risk Management in Financial Institutions for Academic Research 金融机构风险管理模型为学术研究提供的启示

arXiv - QuantFin - Risk Management Pub Date : 2024-06-20 DOI: arxiv-2406.14776

Mahmood Alaghmandan, Olga Streltchenko

引用次数: 0

Robust convex risk measures 稳健的凸风险度量

arXiv - QuantFin - Risk Management Pub Date : 2024-06-18 DOI: arxiv-2406.12999

Marcelo Righi

引用次数: 0

Application of Natural Language Processing in Financial Risk Detection 自然语言处理在金融风险检测中的应用

arXiv - QuantFin - Risk Management Pub Date : 2024-06-14 DOI: arxiv-2406.09765

Liyang Wang, Yu Cheng, Ao Xiang, Jingyu Zhang, Haowei Yang

引用次数: 0

A Multi-step Approach for Minimizing Risk in Decentralized Exchanges 分散式交易所风险最小化的多步骤方法

arXiv - QuantFin - Risk Management Pub Date : 2024-06-11 DOI: arxiv-2406.07200

Daniele Maria Di Nosse, Federico Gatta

{"title":"A Multi-step Approach for Minimizing Risk in Decentralized Exchanges","authors":"Daniele Maria Di Nosse, Federico Gatta","doi":"arxiv-2406.07200","DOIUrl":"https://doi.org/arxiv-2406.07200","url":null,"abstract":"Decentralized Exchanges are becoming even more predominant in today's\u0000finance. Driven by the need to study this phenomenon from an academic\u0000perspective, the SIAG/FME Code Quest 2023 was announced. Specifically,\u0000participating teams were asked to implement, in Python, the basic functions of\u0000an Automated Market Maker and a liquidity provision strategy in an Automated\u0000Market Maker to minimize the Conditional Value at Risk, a critical measure of\u0000investment risk. As the competition's winning team, we highlight our approach\u0000in this work. In particular, as the dependence of the final return on the\u0000initial wealth distribution is highly non-linear, we cannot use standard ad-hoc\u0000approaches. Additionally, classical minimization techniques would require a\u0000significant computational load due to the cost of the target function. For\u0000these reasons, we propose a three-step approach. In the first step, the target\u0000function is approximated by a Kernel Ridge Regression. Then, the approximating\u0000function is minimized. In the final step, the previously discovered minimum is\u0000utilized as the starting point for directly optimizing the desired target\u0000function. By using this procedure, we can both reduce the computational\u0000complexity and increase the accuracy of the solution. Finally, the overall\u0000computational load is further reduced thanks to an algorithmic trick concerning\u0000the returns simulation and the usage of Cython.","PeriodicalId":501128,"journal":{"name":"arXiv - QuantFin - Risk Management","volume":"111 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141508373","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

Adaptive combinations of tail-risk forecasts 尾部风险预测的自适应组合

arXiv - QuantFin - Risk Management Pub Date : 2024-06-10 DOI: arxiv-2406.06235

Alessandra Amendola, Vincenzo Candila, Antonio Naimoli, Giuseppe Storti

{"title":"Adaptive combinations of tail-risk forecasts","authors":"Alessandra Amendola, Vincenzo Candila, Antonio Naimoli, Giuseppe Storti","doi":"arxiv-2406.06235","DOIUrl":"https://doi.org/arxiv-2406.06235","url":null,"abstract":"In order to meet the increasingly stringent global standards of banking\u0000management and regulation, several methods have been proposed in the literature\u0000for forecasting tail risk measures such as the Value-at-Risk (VaR) and Expected\u0000Shortfall (ES). However, regardless of the approach used, there are several\u0000sources of uncertainty, including model specifications, data-related issues and\u0000the estimation procedure, which can significantly affect the accuracy of VaR\u0000and ES measures. Aiming to mitigate the influence of these sources of\u0000uncertainty and improve the predictive performance of individual models, we\u0000propose novel forecast combination strategies based on the Model Confidence Set\u0000(MCS). In particular, consistent joint VaR and ES loss functions within the MCS\u0000framework are used to adaptively combine forecasts generated by a wide range of\u0000parametric, semi-parametric, and non-parametric models. Our results reveal that\u0000the proposed combined predictors provide a suitable alternative for forecasting\u0000risk measures, passing the usual backtests, entering the set of superior models\u0000of the MCS, and usually exhibiting lower standard deviations than other model\u0000specifications.","PeriodicalId":501128,"journal":{"name":"arXiv - QuantFin - Risk Management","volume":"45 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141528272","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

Advancing Anomaly Detection: Non-Semantic Financial Data Encoding with LLMs 推进异常检测：使用 LLM 进行非语义金融数据编码

arXiv - QuantFin - Risk Management Pub Date : 2024-06-05 DOI: arxiv-2406.03614

Alexander BakumenkoClemson University, USA, Kateřina Hlaváčková-SchindlerUniversity of Vienna, Austria, Claudia PlantUniversity of Vienna, Austria, Nina C. HubigClemson University, USA

{"title":"Advancing Anomaly Detection: Non-Semantic Financial Data Encoding with LLMs","authors":"Alexander BakumenkoClemson University, USA, Kateřina Hlaváčková-SchindlerUniversity of Vienna, Austria, Claudia PlantUniversity of Vienna, Austria, Nina C. HubigClemson University, USA","doi":"arxiv-2406.03614","DOIUrl":"https://doi.org/arxiv-2406.03614","url":null,"abstract":"Detecting anomalies in general ledger data is of utmost importance to ensure\u0000trustworthiness of financial records. Financial audits increasingly rely on\u0000machine learning (ML) algorithms to identify irregular or potentially\u0000fraudulent journal entries, each characterized by a varying number of\u0000transactions. In machine learning, heterogeneity in feature dimensions adds\u0000significant complexity to data analysis. In this paper, we introduce a novel\u0000approach to anomaly detection in financial data using Large Language Models\u0000(LLMs) embeddings. To encode non-semantic categorical data from real-world\u0000financial records, we tested 3 pre-trained general purpose sentence-transformer\u0000models. For the downstream classification task, we implemented and evaluated 5\u0000optimized ML models including Logistic Regression, Random Forest, Gradient\u0000Boosting Machines, Support Vector Machines, and Neural Networks. Our\u0000experiments demonstrate that LLMs contribute valuable information to anomaly\u0000detection as our models outperform the baselines, in selected settings even by\u0000a large margin. The findings further underscore the effectiveness of LLMs in\u0000enhancing anomaly detection in financial journal entries, particularly by\u0000tackling feature sparsity. We discuss a promising perspective on using LLM\u0000embeddings for non-semantic data in the financial context and beyond.","PeriodicalId":501128,"journal":{"name":"arXiv - QuantFin - Risk Management","volume":"67 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141548173","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

Analisis cuantitativo de riesgos utilizando "MCSimulRisk" como herramienta didactica 将 "MCSimulRisk "作为教学工具进行定量风险分析

arXiv - QuantFin - Risk Management Pub Date : 2024-05-31 DOI: arxiv-2405.20688

Fernando Acebes, David Curto, Juan de Anton, Felix Villafanez

引用次数: 0

Project Risk Management from the bottom-up: Activity Risk Index 自下而上的项目风险管理：活动风险指数

arXiv - QuantFin - Risk Management Pub Date : 2024-05-31 DOI: arxiv-2406.00078

Fernando Acebes, Javier Pajares, Jose M Gonzalez-Varona, Adolfo Lopez-Paredes

引用次数: 0