In this study, we constructed daily high-frequency sentiment data and used the VAR method to attempt to predict the next day's implied volatility surface. We utilized 630,000 text data entries from the East Money Stock Forum from 2014 to 2023 and employed deep learning methods such as BERT and LSTM to build daily market sentiment indicators. By applying FFT and EMD methods for sentiment decomposition, we found that high-frequency sentiment had a stronger correlation with at-the-money (ATM) options' implied volatility, while low-frequency sentiment was more strongly correlated with deep out-of-the-money (DOTM) options' implied volatility. Further analysis revealed that the shape of the implied volatility surface contains richer market sentiment information beyond just market panic. We demonstrated that incorporating this sentiment information can improve the accuracy of implied volatility surface predictions.
{"title":"Degree of Irrationality: Sentiment and Implied Volatility Surface","authors":"Jiahao Weng, Yan Xie","doi":"arxiv-2405.11730","DOIUrl":"https://doi.org/arxiv-2405.11730","url":null,"abstract":"In this study, we constructed daily high-frequency sentiment data and used\u0000the VAR method to attempt to predict the next day's implied volatility surface.\u0000We utilized 630,000 text data entries from the East Money Stock Forum from 2014\u0000to 2023 and employed deep learning methods such as BERT and LSTM to build daily\u0000market sentiment indicators. By applying FFT and EMD methods for sentiment\u0000decomposition, we found that high-frequency sentiment had a stronger\u0000correlation with at-the-money (ATM) options' implied volatility, while\u0000low-frequency sentiment was more strongly correlated with deep out-of-the-money\u0000(DOTM) options' implied volatility. Further analysis revealed that the shape of\u0000the implied volatility surface contains richer market sentiment information\u0000beyond just market panic. We demonstrated that incorporating this sentiment\u0000information can improve the accuracy of implied volatility surface predictions.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141148252","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}
Carlos Alberto Durigan Junior, Mauro De Mesquita Spinola, Rodrigo Franco Gonçalves, Fernando José Barbin Laurindo
Central Bank Digital Currency (CBDC) can be defined as a virtual currency based on node network and digital encryption algorithm issued by a country which has a legal credit protection. CBDCs are supported by Distributed Ledger Technologies (DLTs), and they may allow a universal means of payments for the digital era. There are many ways to proceed, they all require central banks to develop technological expertise. Considering these points, it is important to understand the new IT governance in the financial markets due to CBDC and digital economy. Information Technology is an essential driver that will allow the new financial industry design. This paper has the objective to answer two questions through an updated Systematic Literature Review (SLR). The first question is What IT resources and tools have been considered or applied to set the governance of CBDC adoption? The second; Identify IT governance models in the financial market due to CBDC adoption. Bank for International Settlements (BIS) publications, Scopus and Web of Science were considered as sources of studies. After the strings and including criteria were applied, fourteen papers were analyzed. This paper finds many IT resources used in the CBDC adoption and some preliminary IT design related to the IT governance of CBDC, in the results and discussion section the findings are more detailed. Finally, limitations and future work are considered. Keywords: Blockchain, Central Bank Digital Currency (CBDC), Digital Economy, Distributed Ledger Technology (DLT), Information Technology (IT), IT governance.
中央银行数字货币（CBDC）可定义为一种基于节点网络和数字加密算法的虚拟货币，由拥有合法信用保护的国家发行。中央银行数字货币由分布式账本技术（DLT）支持，可以成为数字时代的通用支付手段。前进的道路有很多，但都需要中央银行发展技术专长。考虑到这些要点，了解因 CBDC 和数字经济而在金融市场中出现的新信息技术管理非常重要。信息技术是新金融业设计的重要驱动力。本文旨在通过最新的系统文献综述（SLR）回答两个问题。第一个问题是，在采用 CBDC 的治理过程中，考虑或应用了哪些 IT 资源和工具？第二个问题是：确定采用 CBDC 后金融市场的 IT 治理模式。研究来源包括国际清算银行（BIS）出版物、Scopus 和 Web of Science。在应用了字符串和收录标准后，对 14 篇论文进行了分析。本文发现了许多应用于 CBDC 的 IT 资源，以及一些与 CBDC IT 治理相关的初步 IT 设计。最后，考虑了局限性和未来工作。关键词：区块链区块链、央行数字货币（CBDC）、数字经济、分布式账本技术（DLT）、信息技术（IT）、IT治理。
{"title":"Central Bank Digital Currency: The Advent of its IT Governance in the financial markets","authors":"Carlos Alberto Durigan Junior, Mauro De Mesquita Spinola, Rodrigo Franco Gonçalves, Fernando José Barbin Laurindo","doi":"arxiv-2407.07898","DOIUrl":"https://doi.org/arxiv-2407.07898","url":null,"abstract":"Central Bank Digital Currency (CBDC) can be defined as a virtual currency\u0000based on node network and digital encryption algorithm issued by a country\u0000which has a legal credit protection. CBDCs are supported by Distributed Ledger\u0000Technologies (DLTs), and they may allow a universal means of payments for the\u0000digital era. There are many ways to proceed, they all require central banks to\u0000develop technological expertise. Considering these points, it is important to\u0000understand the new IT governance in the financial markets due to CBDC and\u0000digital economy. Information Technology is an essential driver that will allow\u0000the new financial industry design. This paper has the objective to answer two\u0000questions through an updated Systematic Literature Review (SLR). The first\u0000question is What IT resources and tools have been considered or applied to set\u0000the governance of CBDC adoption? The second; Identify IT governance models in\u0000the financial market due to CBDC adoption. Bank for International Settlements\u0000(BIS) publications, Scopus and Web of Science were considered as sources of\u0000studies. After the strings and including criteria were applied, fourteen papers\u0000were analyzed. This paper finds many IT resources used in the CBDC adoption and\u0000some preliminary IT design related to the IT governance of CBDC, in the results\u0000and discussion section the findings are more detailed. Finally, limitations and\u0000future work are considered. Keywords: Blockchain, Central Bank Digital Currency\u0000(CBDC), Digital Economy, Distributed Ledger Technology (DLT), Information\u0000Technology (IT), IT governance.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141609742","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}
Gen Norman Thomas, Siti Mutiara Ramadhanti Nur, Lely Indriaty
This study aims to analyze the impact of financial literacy, social capital and financial technology on financial inclusion. The research method used a quantitative research method, in which questionnaires were distributed to 100 active students in the economics faculty at 7 private colleges in Tangerang, Indonesia. Based on the results of data processing using SPSS version 23, it results that financial literacy, social capital and financial technology partially have a positive and significant influence on financial inclusion. The results of this study provide input that financial literacy needs to be increased because it is not yet equivalent to financial inclusion, and reducing the gap between financial literacy and financial inclusion is only 2.74%. Another benefit of this research is to give an understanding to students that students should be independent actors or users of financial technology products and that students should become pioneers in delivering financial knowledge, financial behavior and financial attitudes to the wider community.
{"title":"The Impact of Financial Literacy, Social Capital, and Financial Technology on Financial Inclusion of Indonesian Students","authors":"Gen Norman Thomas, Siti Mutiara Ramadhanti Nur, Lely Indriaty","doi":"arxiv-2405.06570","DOIUrl":"https://doi.org/arxiv-2405.06570","url":null,"abstract":"This study aims to analyze the impact of financial literacy, social capital\u0000and financial technology on financial inclusion. The research method used a\u0000quantitative research method, in which questionnaires were distributed to 100\u0000active students in the economics faculty at 7 private colleges in Tangerang,\u0000Indonesia. Based on the results of data processing using SPSS version 23, it\u0000results that financial literacy, social capital and financial technology\u0000partially have a positive and significant influence on financial inclusion. The\u0000results of this study provide input that financial literacy needs to be\u0000increased because it is not yet equivalent to financial inclusion, and reducing\u0000the gap between financial literacy and financial inclusion is only 2.74%.\u0000Another benefit of this research is to give an understanding to students that\u0000students should be independent actors or users of financial technology products\u0000and that students should become pioneers in delivering financial knowledge,\u0000financial behavior and financial attitudes to the wider community.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140929300","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}
Entropy is a very useful concept from physics that tries to explain how a system behaves from a point of view of the thermodynamics. However, there are two ways to explain entropy, and it depends on if we are studying a microsystem or a microsystem. From a macroscopically point of view, it is important to describe if the system is a reversible system or not. However, form the microscopically point of view, the concept of chaos is related to entropy. In such case, entropy measures the amount of disorder into the system. Otherwise, the idea of connecting at the same time the analysis of the macro and micro system with the use of entropy it is not very common.
{"title":"Entropy and Economics","authors":"Martin Pomares Calero","doi":"arxiv-2407.00022","DOIUrl":"https://doi.org/arxiv-2407.00022","url":null,"abstract":"Entropy is a very useful concept from physics that tries to explain how a\u0000system behaves from a point of view of the thermodynamics. However, there are\u0000two ways to explain entropy, and it depends on if we are studying a microsystem\u0000or a microsystem. From a macroscopically point of view, it is important to\u0000describe if the system is a reversible system or not. However, form the\u0000microscopically point of view, the concept of chaos is related to entropy. In\u0000such case, entropy measures the amount of disorder into the system. Otherwise,\u0000the idea of connecting at the same time the analysis of the macro and micro\u0000system with the use of entropy it is not very common.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141509538","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}
This paper introduces a novel methodology for index return forecasting, blending highly correlated stock prices, advanced deep learning techniques, and intricate factor integration. Departing from conventional cap-weighted approaches, our innovative framework promises to reimagine traditional methodologies, offering heightened diversification, amplified performance capture, and nuanced market depiction. At its core lies the intricate identification of highly correlated company clusters, fueling predictive accuracy and robustness. By harnessing these interconnected constellations, we unlock a profound comprehension of market dynamics, bestowing both investment entities and individual enterprises with invaluable performance insights. Moreover, our methodology integrates pivotal factors such as indexes and ETFs, seamlessly woven with Hierarchical Risk Parity (HRP) portfolio optimization, to elevate performance and fortify risk management. This comprehensive amalgamation refines risk diversification, fortifying portfolio resilience against turbulent market forces. The implications reverberate resoundingly. Investment entities stand poised to calibrate against competitors with surgical precision, tactically sidestepping industry-specific pitfalls, and sculpting bespoke investment strategies to capitalize on market fluctuations. Concurrently, individual enterprises find empowerment in aligning strategic endeavors with market trajectories, discerning key competitors, and navigating volatility with steadfast resilience. In essence, this research marks a pivotal moment in economic discourse, unveiling novel methodologies poised to redefine decision-making paradigms and elevate performance benchmarks for both investment entities and individual enterprises navigating the intricate tapestry of financial realms.
{"title":"Transforming Investment Strategies and Strategic Decision-Making: Unveiling a Novel Methodology for Enhanced Performance and Risk Management in Financial Markets","authors":"Tian Tian, Ricky Cooper, Jiahao Deng, Qingquan Zhang","doi":"arxiv-2405.01892","DOIUrl":"https://doi.org/arxiv-2405.01892","url":null,"abstract":"This paper introduces a novel methodology for index return forecasting,\u0000blending highly correlated stock prices, advanced deep learning techniques, and\u0000intricate factor integration. Departing from conventional cap-weighted\u0000approaches, our innovative framework promises to reimagine traditional\u0000methodologies, offering heightened diversification, amplified performance\u0000capture, and nuanced market depiction. At its core lies the intricate\u0000identification of highly correlated company clusters, fueling predictive\u0000accuracy and robustness. By harnessing these interconnected constellations, we\u0000unlock a profound comprehension of market dynamics, bestowing both investment\u0000entities and individual enterprises with invaluable performance insights.\u0000Moreover, our methodology integrates pivotal factors such as indexes and ETFs,\u0000seamlessly woven with Hierarchical Risk Parity (HRP) portfolio optimization, to\u0000elevate performance and fortify risk management. This comprehensive\u0000amalgamation refines risk diversification, fortifying portfolio resilience\u0000against turbulent market forces. The implications reverberate resoundingly.\u0000Investment entities stand poised to calibrate against competitors with surgical\u0000precision, tactically sidestepping industry-specific pitfalls, and sculpting\u0000bespoke investment strategies to capitalize on market fluctuations.\u0000Concurrently, individual enterprises find empowerment in aligning strategic\u0000endeavors with market trajectories, discerning key competitors, and navigating\u0000volatility with steadfast resilience. In essence, this research marks a pivotal\u0000moment in economic discourse, unveiling novel methodologies poised to redefine\u0000decision-making paradigms and elevate performance benchmarks for both\u0000investment entities and individual enterprises navigating the intricate\u0000tapestry of financial realms.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140929306","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}
Recently, many works have proposed various financial large language models (FinLLMs) by pre-training from scratch or fine-tuning open-sourced LLMs on financial corpora. However, existing FinLLMs exhibit unsatisfactory performance in understanding financial text when numeric variables are involved in questions. In this paper, we propose a novel LLM, called numeric-sensitive large language model (NumLLM), for Chinese finance. We first construct a financial corpus from financial textbooks which is essential for improving numeric capability of LLMs during fine-tuning. After that, we train two individual low-rank adaptation (LoRA) modules by fine-tuning on our constructed financial corpus. One module is for adapting general-purpose LLMs to financial domain, and the other module is for enhancing the ability of NumLLM to understand financial text with numeric variables. Lastly, we merge the two LoRA modules into the foundation model to obtain NumLLM for inference. Experiments on financial question-answering benchmark show that NumLLM can boost the performance of the foundation model and can achieve the best overall performance compared to all baselines, on both numeric and non-numeric questions.
{"title":"NumLLM: Numeric-Sensitive Large Language Model for Chinese Finance","authors":"Huan-Yi Su, Ke Wu, Yu-Hao Huang, Wu-Jun Li","doi":"arxiv-2405.00566","DOIUrl":"https://doi.org/arxiv-2405.00566","url":null,"abstract":"Recently, many works have proposed various financial large language models\u0000(FinLLMs) by pre-training from scratch or fine-tuning open-sourced LLMs on\u0000financial corpora. However, existing FinLLMs exhibit unsatisfactory performance\u0000in understanding financial text when numeric variables are involved in\u0000questions. In this paper, we propose a novel LLM, called numeric-sensitive\u0000large language model (NumLLM), for Chinese finance. We first construct a\u0000financial corpus from financial textbooks which is essential for improving\u0000numeric capability of LLMs during fine-tuning. After that, we train two\u0000individual low-rank adaptation (LoRA) modules by fine-tuning on our constructed\u0000financial corpus. One module is for adapting general-purpose LLMs to financial\u0000domain, and the other module is for enhancing the ability of NumLLM to\u0000understand financial text with numeric variables. Lastly, we merge the two LoRA\u0000modules into the foundation model to obtain NumLLM for inference. Experiments\u0000on financial question-answering benchmark show that NumLLM can boost the\u0000performance of the foundation model and can achieve the best overall\u0000performance compared to all baselines, on both numeric and non-numeric\u0000questions.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140830407","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}
Claudio Bellei, Muhua Xu, Ross Phillips, Tom Robinson, Mark Weber, Tim Kaler, Charles E. Leiserson, Arvind, Jie Chen
Subgraph representation learning is a technique for analyzing local structures (or shapes) within complex networks. Enabled by recent developments in scalable Graph Neural Networks (GNNs), this approach encodes relational information at a subgroup level (multiple connected nodes) rather than at a node level of abstraction. We posit that certain domain applications, such as anti-money laundering (AML), are inherently subgraph problems and mainstream graph techniques have been operating at a suboptimal level of abstraction. This is due in part to the scarcity of annotated datasets of real-world size and complexity, as well as the lack of software tools for managing subgraph GNN workflows at scale. To enable work in fundamental algorithms as well as domain applications in AML and beyond, we introduce Elliptic2, a large graph dataset containing 122K labeled subgraphs of Bitcoin clusters within a background graph consisting of 49M node clusters and 196M edge transactions. The dataset provides subgraphs known to be linked to illicit activity for learning the set of "shapes" that money laundering exhibits in cryptocurrency and accurately classifying new criminal activity. Along with the dataset we share our graph techniques, software tooling, promising early experimental results, and new domain insights already gleaned from this approach. Taken together, we find immediate practical value in this approach and the potential for a new standard in anti-money laundering and forensic analytics in cryptocurrencies and other financial networks.
{"title":"The Shape of Money Laundering: Subgraph Representation Learning on the Blockchain with the Elliptic2 Dataset","authors":"Claudio Bellei, Muhua Xu, Ross Phillips, Tom Robinson, Mark Weber, Tim Kaler, Charles E. Leiserson, Arvind, Jie Chen","doi":"arxiv-2404.19109","DOIUrl":"https://doi.org/arxiv-2404.19109","url":null,"abstract":"Subgraph representation learning is a technique for analyzing local\u0000structures (or shapes) within complex networks. Enabled by recent developments\u0000in scalable Graph Neural Networks (GNNs), this approach encodes relational\u0000information at a subgroup level (multiple connected nodes) rather than at a\u0000node level of abstraction. We posit that certain domain applications, such as\u0000anti-money laundering (AML), are inherently subgraph problems and mainstream\u0000graph techniques have been operating at a suboptimal level of abstraction. This\u0000is due in part to the scarcity of annotated datasets of real-world size and\u0000complexity, as well as the lack of software tools for managing subgraph GNN\u0000workflows at scale. To enable work in fundamental algorithms as well as domain\u0000applications in AML and beyond, we introduce Elliptic2, a large graph dataset\u0000containing 122K labeled subgraphs of Bitcoin clusters within a background graph\u0000consisting of 49M node clusters and 196M edge transactions. The dataset\u0000provides subgraphs known to be linked to illicit activity for learning the set\u0000of \"shapes\" that money laundering exhibits in cryptocurrency and accurately\u0000classifying new criminal activity. Along with the dataset we share our graph\u0000techniques, software tooling, promising early experimental results, and new\u0000domain insights already gleaned from this approach. Taken together, we find\u0000immediate practical value in this approach and the potential for a new standard\u0000in anti-money laundering and forensic analytics in cryptocurrencies and other\u0000financial networks.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140830274","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}
Jiayue Zhang, Ken Seng Tan, Tony S. Wirjanto, Lysa Porth
This paper extends the application of ESG score assessment methodologies from large corporations to individual farmers' production, within the context of climate change. Our proposal involves the integration of crucial agricultural sustainability variables into conventional personal credit evaluation frameworks, culminating in the formulation of a holistic sustainable credit rating referred to as the Environmental, Social, Economics (ESE) score. This ESE score is integrated into theoretical joint liability models, to gain valuable insights into optimal group sizes and individual-ESE score relationships. Additionally, we adopt a mean-variance utility function for farmers to effectively capture the risk associated with anticipated profits. Through a set of simulation exercises, the paper investigates the implications of incorporating ESE scores into credit evaluation systems, offering a nuanced comprehension of the repercussions under various climatic conditions.
本文在气候变化的背景下，将环境、社会和经济评分评估方法的应用范围从大型企业扩展到个体农民的生产。我们的建议包括将关键的农业可持续发展变量纳入传统的个人信用评估框架，最终形成一个整体的可持续信用评级，称为环境、社会和经济（ESE）评分。该 ESE 分数被纳入理论上的连带责任模型中，以获得关于最佳群体规模和个人与 ESE 分数关系的宝贵见解。此外，我们还采用了农民的均值-方差效用函数，以有效捕捉与预期利润相关的风险。通过一系列模拟练习，本文研究了将 ESE 分数纳入信贷评估系统的影响，并对各种气候条件下的影响进行了细致的分析。
{"title":"Joint Liability Model with Adaptation to Climate Change","authors":"Jiayue Zhang, Ken Seng Tan, Tony S. Wirjanto, Lysa Porth","doi":"arxiv-2404.13818","DOIUrl":"https://doi.org/arxiv-2404.13818","url":null,"abstract":"This paper extends the application of ESG score assessment methodologies from\u0000large corporations to individual farmers' production, within the context of\u0000climate change. Our proposal involves the integration of crucial agricultural\u0000sustainability variables into conventional personal credit evaluation\u0000frameworks, culminating in the formulation of a holistic sustainable credit\u0000rating referred to as the Environmental, Social, Economics (ESE) score. This\u0000ESE score is integrated into theoretical joint liability models, to gain\u0000valuable insights into optimal group sizes and individual-ESE score\u0000relationships. Additionally, we adopt a mean-variance utility function for\u0000farmers to effectively capture the risk associated with anticipated profits.\u0000Through a set of simulation exercises, the paper investigates the implications\u0000of incorporating ESE scores into credit evaluation systems, offering a nuanced\u0000comprehension of the repercussions under various climatic conditions.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140799211","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}
The progress of humanity is driven by those successful discoveries accompanied by countless failed experiments. Researchers often seek the potential research directions by reading and then verifying them through experiments. The process imposes a significant burden on researchers. In the past decade, the data-driven black-box deep learning method demonstrates its effectiveness in a wide range of real-world scenarios, which exacerbates the experimental burden of researchers and thus renders the potential successful discoveries veiled. Therefore, automating such a research and development (R&D) process is an urgent need. In this paper, we serve as the first effort to formalize the goal by proposing a Real-world Data-centric automatic R&D Benchmark, namely RD2Bench. RD2Bench benchmarks all the operations in data-centric automatic R&D (D-CARD) as a whole to navigate future work toward our goal directly. We focuses on evaluating the interaction and synergistic effects of various model capabilities and aiding to select the well-performed trustworthy models. Although RD2Bench is very challenging to the state-of-the-art (SOTA) large language model (LLM) named GPT-4, indicating ample research opportunities and more research efforts, LLMs possess promising potential to bring more significant development to D-CARD: They are able to implement some simple methods without adopting any additional techniques. We appeal to future work to take developing techniques for tackling automatic R&D into consideration, thus bringing the opportunities of the potential revolutionary upgrade to human productivity.
{"title":"RD2Bench: Toward Data-Centric Automatic R&D","authors":"Haotian Chen, Xinjie Shen, Zeqi Ye, Xiao Yang, Xu Yang, Weiqing Liu, Jiang Bian","doi":"arxiv-2404.11276","DOIUrl":"https://doi.org/arxiv-2404.11276","url":null,"abstract":"The progress of humanity is driven by those successful discoveries\u0000accompanied by countless failed experiments. Researchers often seek the\u0000potential research directions by reading and then verifying them through\u0000experiments. The process imposes a significant burden on researchers. In the\u0000past decade, the data-driven black-box deep learning method demonstrates its\u0000effectiveness in a wide range of real-world scenarios, which exacerbates the\u0000experimental burden of researchers and thus renders the potential successful\u0000discoveries veiled. Therefore, automating such a research and development (R&D)\u0000process is an urgent need. In this paper, we serve as the first effort to\u0000formalize the goal by proposing a Real-world Data-centric automatic R&D\u0000Benchmark, namely RD2Bench. RD2Bench benchmarks all the operations in\u0000data-centric automatic R&D (D-CARD) as a whole to navigate future work toward\u0000our goal directly. We focuses on evaluating the interaction and synergistic\u0000effects of various model capabilities and aiding to select the well-performed\u0000trustworthy models. Although RD2Bench is very challenging to the\u0000state-of-the-art (SOTA) large language model (LLM) named GPT-4, indicating\u0000ample research opportunities and more research efforts, LLMs possess promising\u0000potential to bring more significant development to D-CARD: They are able to\u0000implement some simple methods without adopting any additional techniques. We\u0000appeal to future work to take developing techniques for tackling automatic R&D\u0000into consideration, thus bringing the opportunities of the potential\u0000revolutionary upgrade to human productivity.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140616511","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}
Total value locked (TVL) is widely used to measure the size and popularity of protocols and the broader ecosystem in decentralized finance (DeFi). However, the prevalent TVL calculation framework suffers from a "double counting" issue that results in an inflated metric. We find existing methodologies addressing double counting either inconsistent or flawed. To mitigate the double counting issue, we formalize the TVL framework and propose a new framework, total value redeemable (TVR), designed to accurately assess the true value within individual DeFi protocol and DeFi systems. The formalization of TVL indicates that decentralized financial contagion propagates through derivative tokens across the complex network of DeFi protocols and escalates liquidations and stablecoin depegging during market turmoil. By mirroring the concept of money multiplier in traditional finance (TradFi), we construct the DeFi multiplier to quantify the double counting in TVL. Our empirical analysis demonstrates a notable enhancement in the performance of TVR relative to TVL. Specifically, during the peak of DeFi activity on December 2, 2021, the discrepancy between TVL and TVR widened to $139.87 billion, resulting in a TVL-to-TVR ratio of approximately 2. We further show that TVR is a more stable metric than TVL, especially during market turmoil. For instance, a 25% decrease in the price of Ether (ETH) results in an overestimation of the DeFi market value by more than $1 billion when measuring using TVL as opposed to TVR. Overall, our findings suggest that TVR provides a more reliable and stable metric compared to the traditional TVL calculation.
{"title":"Piercing the Veil of TVL: DeFi Reappraised","authors":"Yichen Luo, Yebo Feng, Jiahua Xu, Paolo Tasca","doi":"arxiv-2404.11745","DOIUrl":"https://doi.org/arxiv-2404.11745","url":null,"abstract":"Total value locked (TVL) is widely used to measure the size and popularity of\u0000protocols and the broader ecosystem in decentralized finance (DeFi). However,\u0000the prevalent TVL calculation framework suffers from a \"double counting\" issue\u0000that results in an inflated metric. We find existing methodologies addressing\u0000double counting either inconsistent or flawed. To mitigate the double counting\u0000issue, we formalize the TVL framework and propose a new framework, total value\u0000redeemable (TVR), designed to accurately assess the true value within\u0000individual DeFi protocol and DeFi systems. The formalization of TVL indicates\u0000that decentralized financial contagion propagates through derivative tokens\u0000across the complex network of DeFi protocols and escalates liquidations and\u0000stablecoin depegging during market turmoil. By mirroring the concept of money\u0000multiplier in traditional finance (TradFi), we construct the DeFi multiplier to\u0000quantify the double counting in TVL. Our empirical analysis demonstrates a\u0000notable enhancement in the performance of TVR relative to TVL. Specifically,\u0000during the peak of DeFi activity on December 2, 2021, the discrepancy between\u0000TVL and TVR widened to $139.87 billion, resulting in a TVL-to-TVR ratio of\u0000approximately 2. We further show that TVR is a more stable metric than TVL,\u0000especially during market turmoil. For instance, a 25% decrease in the price of\u0000Ether (ETH) results in an overestimation of the DeFi market value by more than\u0000$1 billion when measuring using TVL as opposed to TVR. Overall, our findings\u0000suggest that TVR provides a more reliable and stable metric compared to the\u0000traditional TVL calculation.","PeriodicalId":501372,"journal":{"name":"arXiv - QuantFin - General Finance","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140626801","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：