arXiv - QuantFin - General Finance最新文献_第3页

Study of the Impact of the Big Data Era on Accounting and Auditing 大数据时代对会计和审计的影响研究

arXiv - QuantFin - General Finance Pub Date : 2024-03-11 DOI: arxiv-2403.07180

Yuxiang Sun, Jingyi Li, Mengdie Lu, Zongying Guo

引用次数: 0

A Heterogeneous Agent Model of Mortgage Servicing: An Income-based Relief Analysis 抵押贷款服务的异质代理模型：基于收入的救济分析

arXiv - QuantFin - General Finance Pub Date : 2024-02-27 DOI: arxiv-2402.17932

Deepeka Garg, Benjamin Patrick Evans, Leo Ardon, Annapoorani Lakshmi Narayanan, Jared Vann, Udari Madhushani, Makada Henry-Nickie, Sumitra Ganesh

{"title":"A Heterogeneous Agent Model of Mortgage Servicing: An Income-based Relief Analysis","authors":"Deepeka Garg, Benjamin Patrick Evans, Leo Ardon, Annapoorani Lakshmi Narayanan, Jared Vann, Udari Madhushani, Makada Henry-Nickie, Sumitra Ganesh","doi":"arxiv-2402.17932","DOIUrl":"https://doi.org/arxiv-2402.17932","url":null,"abstract":"Mortgages account for the largest portion of household debt in the United\u0000States, totaling around $12 trillion nationwide. In times of financial\u0000hardship, alleviating mortgage burdens is essential for supporting affected\u0000households. The mortgage servicing industry plays a vital role in offering this\u0000assistance, yet there has been limited research modelling the complex\u0000relationship between households and servicers. To bridge this gap, we developed\u0000an agent-based model that explores household behavior and the effectiveness of\u0000relief measures during financial distress. Our model represents households as adaptive learning agents with realistic\u0000financial attributes. These households experience exogenous income shocks,\u0000which may influence their ability to make mortgage payments. Mortgage servicers\u0000provide relief options to these households, who then choose the most suitable\u0000relief based on their unique financial circumstances and individual\u0000preferences. We analyze the impact of various external shocks and the success\u0000of different mortgage relief strategies on specific borrower subgroups. Through this analysis, we show that our model can not only replicate\u0000real-world mortgage studies but also act as a tool for conducting a broad range\u0000of what-if scenario analyses. Our approach offers fine-grained insights that\u0000can inform the development of more effective and inclusive mortgage relief\u0000solutions.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140005623","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

Securing Transactions: A Hybrid Dependable Ensemble Machine Learning Model using IHT-LR and Grid Search 确保交易安全：使用 IHT-LR 和网格搜索的混合可靠集合机器学习模型

arXiv - QuantFin - General Finance Pub Date : 2024-02-22 DOI: arxiv-2402.14389

Md. Alamin Talukder, Rakib Hossen, Md Ashraf Uddin, Mohammed Nasir Uddin, Uzzal Kumar Acharjee

{"title":"Securing Transactions: A Hybrid Dependable Ensemble Machine Learning Model using IHT-LR and Grid Search","authors":"Md. Alamin Talukder, Rakib Hossen, Md Ashraf Uddin, Mohammed Nasir Uddin, Uzzal Kumar Acharjee","doi":"arxiv-2402.14389","DOIUrl":"https://doi.org/arxiv-2402.14389","url":null,"abstract":"Financial institutions and businesses face an ongoing challenge from\u0000fraudulent transactions, prompting the need for effective detection methods.\u0000Detecting credit card fraud is crucial for identifying and preventing\u0000unauthorized transactions.Timely detection of fraud enables investigators to\u0000take swift actions to mitigate further losses. However, the investigation\u0000process is often time-consuming, limiting the number of alerts that can be\u0000thoroughly examined each day. Therefore, the primary objective of a fraud\u0000detection model is to provide accurate alerts while minimizing false alarms and\u0000missed fraud cases. In this paper, we introduce a state-of-the-art hybrid\u0000ensemble (ENS) dependable Machine learning (ML) model that intelligently\u0000combines multiple algorithms with proper weighted optimization using Grid\u0000search, including Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor\u0000(KNN), and Multilayer Perceptron (MLP), to enhance fraud identification. To\u0000address the data imbalance issue, we employ the Instant Hardness Threshold\u0000(IHT) technique in conjunction with Logistic Regression (LR), surpassing\u0000conventional approaches. Our experiments are conducted on a publicly available\u0000credit card dataset comprising 284,807 transactions. The proposed model\u0000achieves impressive accuracy rates of 99.66%, 99.73%, 98.56%, and 99.79%, and a\u0000perfect 100% for the DT, RF, KNN, MLP and ENS models, respectively. The hybrid\u0000ensemble model outperforms existing works, establishing a new benchmark for\u0000detecting fraudulent transactions in high-frequency scenarios. The results\u0000highlight the effectiveness and reliability of our approach, demonstrating\u0000superior performance metrics and showcasing its exceptional potential for\u0000real-world fraud detection applications.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"75 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139949019","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

Regulating Cryptocurrency and Decentralized Finance for an Inclusive Economy 监管加密货币和去中心化金融，实现包容性经济

arXiv - QuantFin - General Finance Pub Date : 2024-02-15 DOI: arxiv-2407.01532

Amrutha Muralidhar, Muralidhar Lakkanna

{"title":"Regulating Cryptocurrency and Decentralized Finance for an Inclusive Economy","authors":"Amrutha Muralidhar, Muralidhar Lakkanna","doi":"arxiv-2407.01532","DOIUrl":"https://doi.org/arxiv-2407.01532","url":null,"abstract":"The evolution of cryptocurrency and decentralized finance (DeFi) marks a\u0000significant shift in the financial landscape, making it more accessible,\u0000inclusive, and participative for various societal groups. However, this\u0000transition from traditional financial institutions to DeFi demands a meticulous\u0000policy framework that strikes a balance between innovation and safeguarding\u0000consumer interests, security, and regulatory compliance. In this script we\u0000explore the imperative need for regulatory frameworks overseeing\u0000cryptocurrencies and DeFi, aiming to leverage their potential for inclusive\u0000economic advancement. It underscores the prevalent challenges within\u0000conventional financial systems, juxtaposing them with the transformative\u0000potential offered by these emergent financial paradigms. By highlighting the\u0000role of robust regulations, we examine their capacity to ensure user security,\u0000fortify market resilience, and spur innovative strides. We aim to proffer\u0000viable strategies for formulating regulatory structures that harmonize the twin\u0000objectives of fostering innovation and upholding fairness within financial\u0000ecosystems.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"227 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141517608","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

Token vs Equity for Startup Financing 代币与股权对初创企业融资的影响

arXiv - QuantFin - General Finance Pub Date : 2024-02-07 DOI: arxiv-2402.04662

Guangye Cao

引用次数: 0

Prioritizing Investments in Cybersecurity: Empirical Evidence from an Event Study on the Determinants of Cyberattack Costs 确定网络安全投资的优先次序：网络攻击成本决定因素事件研究的经验证据

arXiv - QuantFin - General Finance Pub Date : 2024-02-07 DOI: arxiv-2402.04773

Daniel Celeny, Loïc Maréchal, Evgueni Rousselot, Alain Mermoud, Mathias Humbert

{"title":"Prioritizing Investments in Cybersecurity: Empirical Evidence from an Event Study on the Determinants of Cyberattack Costs","authors":"Daniel Celeny, Loïc Maréchal, Evgueni Rousselot, Alain Mermoud, Mathias Humbert","doi":"arxiv-2402.04773","DOIUrl":"https://doi.org/arxiv-2402.04773","url":null,"abstract":"Along with the increasing frequency and severity of cyber incidents,\u0000understanding their economic implications is paramount. In this context, listed\u0000firms' reactions to cyber incidents are compelling to study since they (i) are\u0000a good proxy to estimate the costs borne by other organizations, (ii) have a\u0000critical position in the economy, and (iii) have their financial information\u0000publicly available. We extract listed firms' cyber incident dates and\u0000characteristics from newswire headlines. We use an event study over 2012--2022,\u0000using a three-day window around events and standard benchmarks. We find that\u0000the magnitude of abnormal returns around cyber incidents is on par with\u0000previous studies using newswire or alternative data to identify cyber\u0000incidents. Conversely, as we adjust the standard errors accounting for\u0000event-induced variance and residual cross-correlation, we find that the\u0000previously claimed significance of abnormal returns vanishes. Given these\u0000results, we run a horse race of specifications, in which we test for the\u0000marginal effects of type of cyber incidents, target firm sector, periods, and\u0000their interactions. Data breaches are the most detrimental incident type with\u0000an average loss of -1.3% or (USD -1.9 billion) over the last decade. The\u0000health sector is the most sensitive to cyber incidents, with an average loss of\u0000-5.21% (or USD -1.2 billion), and even more so when these are data breaches.\u0000Instead, we cannot show any time-varying effect of cyber incidents or a\u0000specific effect of the type of news as had previously been advocated.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"16 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139769463","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

A Survey of Large Language Models in Finance (FinLLMs) 金融领域大型语言模型（FinLLMs）概览

arXiv - QuantFin - General Finance Pub Date : 2024-02-04 DOI: arxiv-2402.02315

Jean Lee, Nicholas Stevens, Soyeon Caren Han, Minseok Song

{"title":"A Survey of Large Language Models in Finance (FinLLMs)","authors":"Jean Lee, Nicholas Stevens, Soyeon Caren Han, Minseok Song","doi":"arxiv-2402.02315","DOIUrl":"https://doi.org/arxiv-2402.02315","url":null,"abstract":"Large Language Models (LLMs) have shown remarkable capabilities across a wide\u0000variety of Natural Language Processing (NLP) tasks and have attracted attention\u0000from multiple domains, including financial services. Despite the extensive\u0000research into general-domain LLMs, and their immense potential in finance,\u0000Financial LLM (FinLLM) research remains limited. This survey provides a\u0000comprehensive overview of FinLLMs, including their history, techniques,\u0000performance, and opportunities and challenges. Firstly, we present a\u0000chronological overview of general-domain Pre-trained Language Models (PLMs)\u0000through to current FinLLMs, including the GPT-series, selected open-source\u0000LLMs, and financial LMs. Secondly, we compare five techniques used across\u0000financial PLMs and FinLLMs, including training methods, training data, and\u0000fine-tuning methods. Thirdly, we summarize the performance evaluations of six\u0000benchmark tasks and datasets. In addition, we provide eight advanced financial\u0000NLP tasks and datasets for developing more sophisticated FinLLMs. Finally, we\u0000discuss the opportunities and the challenges facing FinLLMs, such as\u0000hallucination, privacy, and efficiency. To support AI research in finance, we\u0000compile a collection of accessible datasets and evaluation benchmarks on\u0000GitHub.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139769356","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

Attention-based Dynamic Multilayer Graph Neural Networks for Loan Default Prediction 基于注意力的动态多层图神经网络用于贷款违约预测

arXiv - QuantFin - General Finance Pub Date : 2024-02-01 DOI: arxiv-2402.00299

Sahab Zandi, Kamesh Korangi, María Óskarsdóttir, Christophe Mues, Cristián Bravo

{"title":"Attention-based Dynamic Multilayer Graph Neural Networks for Loan Default Prediction","authors":"Sahab Zandi, Kamesh Korangi, María Óskarsdóttir, Christophe Mues, Cristián Bravo","doi":"arxiv-2402.00299","DOIUrl":"https://doi.org/arxiv-2402.00299","url":null,"abstract":"Whereas traditional credit scoring tends to employ only individual borrower-\u0000or loan-level predictors, it has been acknowledged for some time that\u0000connections between borrowers may result in default risk propagating over a\u0000network. In this paper, we present a model for credit risk assessment\u0000leveraging a dynamic multilayer network built from a Graph Neural Network and a\u0000Recurrent Neural Network, each layer reflecting a different source of network\u0000connection. We test our methodology in a behavioural credit scoring context\u0000using a dataset provided by U.S. mortgage financier Freddie Mac, in which\u0000different types of connections arise from the geographical location of the\u0000borrower and their choice of mortgage provider. The proposed model considers\u0000both types of connections and the evolution of these connections over time. We\u0000enhance the model by using a custom attention mechanism that weights the\u0000different time snapshots according to their importance. After testing multiple\u0000configurations, a model with GAT, LSTM, and the attention mechanism provides\u0000the best results. Empirical results demonstrate that, when it comes to\u0000predicting probability of default for the borrowers, our proposed model brings\u0000both better results and novel insights for the analysis of the importance of\u0000connections and timestamps, compared to traditional methods.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139665914","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

Cash and Card Acceptance in Retail Payments: Motivations and Factors 零售支付中的现金和银行卡接受度：动机和因素

arXiv - QuantFin - General Finance Pub Date : 2024-01-15 DOI: arxiv-2401.07682

Samuel Vandak, Geoffrey Goodell

{"title":"Cash and Card Acceptance in Retail Payments: Motivations and Factors","authors":"Samuel Vandak, Geoffrey Goodell","doi":"arxiv-2401.07682","DOIUrl":"https://doi.org/arxiv-2401.07682","url":null,"abstract":"The landscape of payment methods in retail is a complex and evolving area.\u0000Vendors are motivated to conduct an appropriate analysis to decide what payment\u0000methods to accept out of a vast range of options. Many factors are included in\u0000this decision process, some qualitative and some quantitative. The following\u0000research project investigates vendors' acceptance of cards and cash from\u0000various viewpoints, all chosen to represent a novel perspective, including the\u0000barriers and preferences for each and correlations with external demographic\u0000factors. We observe that lower interchange fees, limited in this instance by\u0000the regulatory framework, play a crucial role in facilitating merchants'\u0000acceptance of card payments. The regulatory constraints on interchange fees\u0000create a favorable cost structure for merchants, making card payment adoption\u0000financially feasible. However, additional factors like technological readiness\u0000and consumer preferences might also play a significant role in their\u0000decision-making process. We also note that aggregate Merchant Service Providers\u0000(MSPs) have positively impacted the payment landscape by offering more\u0000competitive fee rates, particularly beneficial for small merchants and\u0000entrepreneurs. However, associated risks, such as account freezes or abrupt\u0000terminations, pose challenges and often lack transparency. Last, the\u0000quantitative analysis of the relationship between demographic variables and\u0000acceptance of payment types is presented. This analysis combines the current\u0000landscape of payment acceptance in the UK with data from the most recent census\u0000from 2021. We show that the unemployment rates shape card and cash acceptance,\u0000age affects contactless preference, and work-from-home impacts credit card\u0000preference.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"39 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139483262","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

Non-Atomic Arbitrage in Decentralized Finance 分散金融中的非原子套利

arXiv - QuantFin - General Finance Pub Date : 2024-01-03 DOI: arxiv-2401.01622

Lioba Heimbach, Vabuk Pahari, Eric Schertenleib

{"title":"Non-Atomic Arbitrage in Decentralized Finance","authors":"Lioba Heimbach, Vabuk Pahari, Eric Schertenleib","doi":"arxiv-2401.01622","DOIUrl":"https://doi.org/arxiv-2401.01622","url":null,"abstract":"The prevalence of maximal extractable value (MEV) in the Ethereum ecosystem\u0000has led to a characterization of the latter as a dark forest. Studies of MEV\u0000have thus far largely been restricted to purely on-chain MEV, i.e., sandwich\u0000attacks, cyclic arbitrage, and liquidations. In this work, we shed light on the\u0000prevalence of non-atomic arbitrage on decentralized exchanges (DEXes) on the\u0000Ethereum blockchain. Importantly, non-atomic arbitrage exploits price\u0000differences between DEXes on the Ethereum blockchain as well as exchanges\u0000outside the Ethereum blockchain (i.e., centralized exchanges or DEXes on other\u0000blockchains). Thus, non-atomic arbitrage is a type of MEV that involves actions\u0000on and off the Ethereum blockchain. In our study of non-atomic arbitrage, we uncover that more than a fourth of\u0000the volume on Ethereum's biggest five DEXes from the merge until 31 October\u00002023 can likely be attributed to this type of MEV. We further highlight that\u0000only eleven searchers are responsible for more than 80% of the identified\u0000non-atomic arbitrage volume sitting at a staggering 137 billion US$ and draw a\u0000connection between the centralization of the block construction market and\u0000non-atomic arbitrage. Finally, we discuss the security implications of these\u0000high-value transactions that account for more than 10% of Ethereum's total\u0000block value and outline possible mitigations.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":"132 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139095454","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