https://arxiv.org/api/ZYpR65MYb8WVEtdw9aass07PV9Q 2026-04-13T02:32:11Z 2963 600 15 http://arxiv.org/abs/2106.04114v3 2022-12-22T07:18:22Z

The task we consider is portfolio construction in a speculative market, a fundamental problem in modern finance. While various empirical works now exist to explore deep learning in finance, the theory side is almost non-existent. In this work, we focus on developing a theoretical framework for understanding the use of data augmentation for deep-learning-based approaches to quantitative finance. The proposed theory clarifies the role and necessity of data augmentation for finance; moreover, our theory implies that a simple algorithm of injecting a random noise of strength $\sqrt{|r_{t-1}|}$ to the observed return $r_{t}$ is better than not injecting any noise and a few other financially irrelevant data augmentation techniques.

2021-06-08T05:26:58Z The full version of our work published at 3rd ACM International Conference on AI in Finance (ICAIF'22) Liu Ziyin Kentaro Minami Kentaro Imajo 10.1145/3533271.3561720 http://arxiv.org/abs/2212.10844v1 2022-12-21T08:36:02Z

As of 2022, greenhouse gases (GHG) emissions reporting and auditing are not yet compulsory for all companies and methodologies of measurement and estimation are not unified. We propose a machine learning-based model to estimate scope 1 and scope 2 GHG emissions of companies not reporting them yet. Our model, specifically designed to be transparent and completely adapted to this use case, is able to estimate emissions for a large universe of companies. It shows good out-of-sample global performances as well as good out-of-sample granular performances when evaluating it by sectors, by countries or by revenues buckets. We also compare our results to those of other providers and find our estimates to be more accurate. Thanks to the proposed explainability tools using Shapley values, our model is fully interpretable, the user being able to understand which factors split explain the GHG emissions for each particular company.

2022-12-21T08:36:02Z Jeremi Assael BNPP CIB GM Lab, MICS Thibaut Heurtebize BNPP CIB GM Lab Laurent Carlier BNPP CIB GM Lab François Soupé http://arxiv.org/abs/2212.10308v1 2022-12-20T14:55:51Z

In this paper, we propose a fully decentralized and smart contract-based insurance protocol. We identify various issues in the Decentralized Finance (DeFi) insurance context and propose a solution to overcome these shortcomings. We introduce an economic model that allows for risk transfer without any external dependencies or centralized intermediaries. In particular, our proposal does not need any sort of subjective claim assessment, community voting or external data providers (oracles). Moreover, it solves the problem of over-insurance and proposes various ways to mitigate the capital inefficiencies usually seen with DeFi collateral. The work takes inspiration from peer-to-peer (P2P) insurance and collateralized debt obligations (CDO). We formally describe the protocol, assess its efficiency and key properties and present a reference implementation. Finally, we address limitations, extensions and ideas for further research.

2022-12-20T14:55:51Z Matthias Nadler Felix Bekemeier Fabian Schär http://arxiv.org/abs/2207.12631v2 2022-12-12T07:55:52Z

Microfinance, despite its significant potential for poverty reduction, is facing sustainability hardships due to high default rates. Although many methods in regular finance can estimate credit scores and default probabilities, these methods are not directly applicable to microfinance due to the following unique characteristics: a) under-explored (developing) areas such as rural Africa do not have sufficient prior loan data for microfinance institutions (MFIs) to establish a credit scoring system; b) microfinance applicants may have difficulty providing sufficient information for MFIs to accurately predict default probabilities; and c) many MFIs use group liability (instead of collateral) to secure repayment. Here, we present a novel control-theoretic model of microfinance that accounts for these characteristics. We construct an algorithm to learn microfinance decision policies that achieve financial inclusion, fairness, social welfare, and sustainability. We characterize the convergence conditions to Pareto-optimum and the convergence speeds. We demonstrate, in numerous real and synthetic datasets, that the proposed method accounts for the complexities induced by group liability to produce robust decisions before sufficient loans are given to establish credit scoring systems and for applicants whose default probability cannot be accurately estimated due to missing information. To the best of our knowledge, this paper is the first to connect microfinance and control theory. We envision that the connection will enable safe learning and control techniques to help modernize microfinance and alleviate poverty.

2022-07-26T03:35:18Z 37 pages, 12 figures Christian Kurniawan Xiyu Deng Adhiraj Chakraborty Assane Gueye Niangjun Chen Yorie Nakahira http://arxiv.org/abs/2212.05734v1 2022-12-12T07:15:15Z

Permissionless blockchains offer an information environment where users can interact privately without fear of censorship. Financial services can be programmatically coded via smart contracts to automate transactions without the need for human intervention or knowing user identity. This new paradigm is known as decentralized finance (DeFi). We investigate Compound (a leading DeFi lending protocol) to show how it works in this novel information environment, who its users are, and what factors determine their participation. On-chain transaction data shows that loan durations are short (31 days on average), and many users borrow to support leveraged investment strategies (yield farming). We show that systemic risk in DeFi arises from concentration and interconnection, and how traditional risk management practices can be challenging for DeFi.

2022-12-12T07:15:15Z Kanis Saengchote http://arxiv.org/abs/2007.05884v9 2022-12-07T20:11:55Z

Economists have mainly focused on human capital accumulation rather than on the causes and consequences of human capital depreciation in late adulthood. To investigate how human capital depreciates over the life cycle, we examine how a newly introduced pension program, the National Rural Pension Scheme, affects cognitive performance in rural China. We find significant adverse effects of access to pension benefits on cognitive functioning among the elderly. We detect the most substantial impact of the program on delayed recall, a cognition measure linked to the onset of dementia. In terms of mechanisms, cognitive deterioration in late adulthood is mediated by a substantial reduction in social engagement, volunteering, and activities fostering mental acuity.

2020-07-12T01:31:13Z 62 pages; 1 figure; 6 tables Plamen Nikolov Md Shahadath Hossain http://arxiv.org/abs/2003.11347v3 2022-12-06T10:51:19Z

We analyze how retail investors price the credit risk of online P"P consumer loans in a reverse auction framework where personal interaction is absent. The explained interest rate variance is considerably larger than in comparable studies using bank loan data. This is unexpected, given the difference in experience and the limited set of information provided to the investor. Factors representing economic status significantly influence lender evaluations of the borrower's credit risk. Previous studies on gender discrimination found mixed results. We show that this is likely due to simplified specifications and find surprising results: male borrowers are charged a higher interest rates, conditional on being married and having children. Overall, our results indicate that retail investors exhibit a strong degree of predictability in this weakly regulated market for online P2P consumer loans.

2020-03-25T12:03:33Z 22 Pages Andreas Dietrich Reto Rey http://arxiv.org/abs/2212.02307v1 2022-12-05T14:35:01Z

We use the Grossman \& Stiglitz (1980) framework to build a reference portfolio for uninformed investors and employ this portfolio to assess the performance of actively managed equity mutual funds. We propose an empirical methodology to construct this reference portfolio using the information on prices and supply. We show that mutual funds provide, on average, an insignificant alpha of 23 basis points per year when considering this portfolio as a reference. With the stock market index as a proxy for the market portfolio, the average fund alpha is negative and highly significant, --128 basis points per year. The results are robust when considering various subsets of funds based on their characteristics and their degree of selectivity. In line with rational expectations equilibrium models considering asymmetrically informed investors and partially revealing equilibrium prices, our study supports that active management adds value for uniformed investors.

2022-12-05T14:35:01Z Finance, In press Radu Burlacu CERAG Patrice Fontaine EUROFIDAI Sonia Jimenez-Garcès http://arxiv.org/abs/2212.00018v1 2022-11-30T09:59:55Z

Our main contribution is that we are using AI to discern the key drivers of variation of ESG mentions in the corporate filings. With AI, we are able to separate "dimensions" along which the corporate management presents their ESG policies to the world. These dimensions are 1) diversity, 2) hazardous materials, and 3) greenhouse gasses. We are also able to identify separate "background" dimensions of unofficial ESG activity in the firms, which provide more color into the firms and their shareholders' thinking about their ESG processes. We then measure investors' response to the ESG activity "factors". The AI techniques presented can assist in building better, more reliable and useful ESG ratings systems.

2022-11-30T09:59:55Z Irene Aldridge Payton Martin http://arxiv.org/abs/2212.11787v1 2022-11-30T02:42:13Z

Carbon neutralization is an urgent task in society because of the global warming threat. And carbon trading is an essential market mechanics to solve carbon reduction targets. Macro carbon price prediction is vital in the useful management and decision-making of the carbon market. We focus on the EU carbon market and we choose oil price, coal price, gas price, and DAX index to be the four market factors in predicting carbon price, and also we select carbon emission targets from Paris Accord as the political factor in the carbon market in terms of the macro view of the carbon price prediction. Thus we use these five factors as inputs to predict the future carbon yearly price in 2030 with the support vector regression models. We use grid search and cross validation to guarantee the prediction performance of our models. We believe this model will have great applications in the macro carbon price prediction.

2022-11-30T02:42:13Z Jinhui Li http://arxiv.org/abs/2211.16151v1 2022-11-29T12:24:55Z

How do business-cycles impact startup-valuations? While several studies explore VC startupecosystems and pre-money valuations, relatively-few delve deeper into the role of macro-level economic factors in influencing those startup deals valuations. Using a dataset of 1,089 venturecapital investments in European Union and European Economic Area markets, this article examines macroeconomic, cyclical and macro-sectoral influences on VC startups pre-money VC valuations. Our findings show that business-cycles impact startup-valuation both directly and indirectly. Beyond DCF factors, startup-valuations are impacted via by business-cycles directly, and via local venture-capital market-size. By using a Structural Equation Model approach, our findings contribute to entrepreneurship and financial-intermediary literature by exploring indirect and endogenous relationship possibilities finding that most determinants are transmission-channels rather than independent drivers. Our findings effectively tie-together startup-valuations, intermediary markets, and macroeconomic determinants.

2022-11-29T12:24:55Z 2022 Entrepreneurial Finance (ENTFIN) Annual Meeting, Sep 2022, Bath, United Kingdom Max Berre Benjamin Le Pendeven http://arxiv.org/abs/2112.11338v3 2022-11-28T17:53:28Z

The U.S. electrical grid has undergone substantial transformation with increased penetration of wind and solar -- forms of variable renewable energy (VRE). Despite the benefits of VRE for decarbonization, it has garnered some controversy for inducing unwanted effects in regional electricity markets. In this study, the role of VRE penetration is examined on the system electricity price and price volatility based on hourly, real-time, historical data from six Independent System Operators (ISOs) in the U.S. using quantile and skew t-distribution regressions. After correcting for temporal effects, we found an increase in VRE penetration is associated with decrease in system electricity price in all ISOs studied. The increase in VRE penetration is associated with decrease in temporal price volatility in five out of six ISOs studied. The relationships are non-linear. These results are consistent with the modern portfolio theory where diverse volatile assets may lead to more stable and less risky portfolios.

2021-11-10T16:50:25Z Olukunle O. Owolabi Toryn L. J. Schafer Georgia E. Smits Sanhita Sengupta Sean E. Ryan Lan Wang David S. Matteson Mila Getmansky Sherman Deborah A. Sunter http://arxiv.org/abs/2212.11779v1 2022-11-19T06:47:01Z

Since its inauguration, ChiNext has gone through three time periods with two different regulation regimes and three different sets of listing day trading restrictions. This paper studies the impact of regulation regimes and listing day trading restrictions on the initial return of ChiNext IPOs. We hypothesize that the initial return of a ChiNext IPO contains the issuers intrinsic value and the investors overreaction. The intrinsic value is represented by the IPOs 21st day return (monthly return), and the difference between the monthly and initial returns (intramonth return) is a proxy of the overreaction. We find that all significant variables for all three returns in all three time periods fall into four categories: pre-listing demand, post-listing demand, market condition and pre-listing issuer value. We observe stark contrasts among variable categories for each of the returns in the three time periods, which reveals an evolution of the investors behavior with regard to the progression of regulation regimes. Based on our findings, we argue that the differences among the levels and determinants of initial return, monthly return (intrinsic value) and intramonth return (overreaction) in different time periods can be largely explained by regulation regime changes along two dimensions: 1) approval vs. registration and 2) listing day trading curbs and return limits. We find that IPO pricing is demand-driven under the approval regime, but value-driven under the registration regime. We further compare the impact of regulation regime changes on ChiNext IPO pricing practice, and propose a future research plan on ChiNext IPO pricing efficiency with policy implication.

2022-11-19T06:47:01Z Qi Deng Lunge Dai Zixin Yang Zhong-guo Zhou Monica Hussein Dingyi Chen Mick Swartz http://arxiv.org/abs/2210.04339v2 2022-11-14T20:43:01Z

Non-Fungible Tokens (NFTs) have recently surged to mainstream attention by allowing the exchange of digital assets via blockchains. NFTs have also been adopted by artists to sell digital art. One of the promises of NFTs is broadening participation to the arts market, a traditionally closed and opaque system, to sustain a wider and more diverse set of artists and collectors. A key sign of this effect would be the disappearance or at least reduction in importance of seller-buyer preferential ties, whereby the success of an artist is strongly dependent on the patronage of a single collector. We investigate NFT art seller-buyer networks considering several galleries and a large set of nearly 40,000 sales for over 230M USD in total volume. We find that NFT art is a highly concentrated market driven by few successful sellers and even fewer systematic buyers. High concentration is present in both the number of sales and, even more strongly, in their priced volume. Furthermore, we show that, while a broader-participation market was present in the early phase of NFT art adoption, preferential ties have dominated during market growth, peak and recent decline. We consistently find that the top buyer accounts on average for over 80% of buys for a given seller. Similar trends apply to buyers and their top seller. We conclude that NFT art constitutes, at the present, a highly concentrated market driven by preferential seller-buyer ties.

2022-10-09T20:09:41Z Giovanni Colavizza http://arxiv.org/abs/2212.11765v1 2022-11-13T11:23:02Z

Environmental, social and governance (ESG) engagement of companies moved into the focus of public attention over recent years. With the requirements of compulsory reporting being implemented and investors incorporating sustainability in their investment decisions, the demand for transparent and reliable ESG ratings is increasing. However, automatic approaches for forecasting ESG ratings have been quite scarce despite the increasing importance of the topic. In this paper, we build a model to predict ESG ratings from news articles using the combination of multivariate timeseries construction and deep learning techniques. A news dataset for about 3,000 US companies together with their ratings is also created and released for training. Through the experimental evaluation we find out that our approach provides accurate results outperforming the state-of-the-art, and can be used in practice to support a manual determination or analysis of ESG ratings.

2022-11-13T11:23:02Z Tanja Aue Adam Jatowt Michael Färber