Quant LetterNo. 99
144 items across 6 sections, as sent to readers on 30 May 2025. Paper titles open their ML-Quant page; ↗ goes to the source.
Quantitative-finance and ML-for-finance preprints from arXiv.
26 items
The study finds that using fractional differentiation in data preparation enhances the forecasting performance of predictive models in financial time series, compared to traditional logarithmic returns.
21 shares3 citations todaySource ↗
The research shows that a neural network trained with just 256 trajectories can outperform the Black & Scholes formula and the Leland model in the Geometric Brownian Motion framework, indicating potential for real-time financial series application.
19 shares1 citation todaySource ↗
The paper presents a Machine Learning framework for pricing derivative products based on volatility surfaces, offering accurate results and faster processing than traditional methods, thus facilitating real-time risk analytics and large-scale scenario analysis.
15 shares1 citation todaySource ↗
The research investigates the effectiveness of buffered Probability of Exceedance strategies in long-term retirement planning through multi-period mean-risk portfolio optimization.
14 shares4 citations todaySource ↗
The study replicates previous research on reference-dependent preferences in the Chinese stock market, finding a less significant role and suggesting the need for customized investment strategies in emerging economies.
13 shares1 citation todaySource ↗
The paper presents a new risk-sharing model that allows for risk distribution among a range of agents, demonstrating the existence of risk-minimizing allocations under specific conditions.
12 shares1 citation todaySource ↗
The research suggests using an agent-based model to create synthetic data for categorizing financial investors by their behavior, emphasizing the difficulties in validating and interpreting machine learning methods.
11 shares2 citations todaySource ↗
The study suggests that AI systems need to be 5-6 times more productive than current automation to finance a universal basic income without additional taxes or job creation.
44 shares1 citation todaySource ↗
The paper introduces Experiential Matrix Theory, a new theory for growth and employment in the AI age, and establishes Alignment Economics as a new field of research.
17 sharesSource ↗
The research uses large language models to create a dataset of capital control measures across 196 countries, contributing to the use of these models in economics.
15 sharesSource ↗
The study uses graph theory to show that political dynasties in the Philippines have become stronger and more interconnected, negatively impacting the Human Development Index scores.
13 shares2 citations todaySource ↗
The article introduces a new method for ranking educational institutions based on applicants' choices and test scores, providing a more customizable alternative to traditional systems.
13 sharesSource ↗
The research shows that defendants with unfavorable juries, especially black defendants, have higher conviction rates, indicating a disparity in the criminal justice system.
13 sharesSource ↗
The paper proposes Academic Research Output Futures and Options, financial tools linked to a Research Output Index, as a potential new funding model for universities and research institutions.
12 sharesSource ↗
The study analyzes the effects of South Korea's heavy and chemical industry policies during the East Asian miracle, finding that these policies spurred growth in targeted industries and benefited downstream users.
12 shares114 citations todaySource ↗
The article introduces a concept of marginal fairness for unbiased decision-making in sectors like insurance and finance, disregarding protected attributes such as race, gender, and religion.
23 shares2 citations todaySource ↗
The paper presents a new robust Q-learning algorithm for continuous state spaces, optimizing for the worst-case scenario, with applications like portfolio optimization.
22 shares3 citations todaySource ↗
The study introduces the Faithful Group Shapley Value (FGSV) method for group-level data valuation, which protects against shell company attacks and ensures accurate computation.
19 shares4 citations todaySource ↗
The article discusses recalibrating binary probabilistic classifiers from a distribution shift perspective, introducing two new methods for conservative results in credit risk assessments.
12 shares1 citation todaySource ↗
The paper presents a mathematical framework modeling the role of Generative AI in job scenarios, showing it enhances human skills rather than replacing them, especially benefiting lower-skilled workers.
12 shares2 citations todaySource ↗
OFA Implications: The Ethereum DeFi sector has seen a shift with 80% of transactions now using private RPCs, emphasizing the impact of Order Flow Auctions on transaction efficiency and quality.
59 shares4 citations todaySource ↗
ARIMA with SVM/LSTM: A study using econometric models, machine learning, and deep learning to predict financial trends for the S&P 500 and Bitcoin emphasizes the importance of well-constructed hybrid models for profitable trading strategies.
18 shares20 citations todaySource ↗
Research shows that the success of eco-restoration strategies in European agriculture depends on farm size, with smaller farms benefiting economically from less pesticide use and hedgerow restoration, while larger farms face challenges.
27 shares3 citations todaySource ↗
A new penalized quantile regression model for predicting equity premium outperforms other methods and provides insights into the relationship between predictors and excess return across different quantiles.
22 sharesSource ↗
A study finds that people tend to lower compensation for workers using AI tools, a trend called AI Penalization, indicating that AI adoption in the workplace could increase worker inequality.
20 shares2 citations todaySource ↗
A new machine learning model uses global news sentiment to predict next-day returns for financial instruments, with sentiment dispersion and article impact being key predictive features.
19 shares2 citations todaySource ↗
Working papers in finance and economics from SSRN.
58 items
Personal investments by Venture Capital partners can negatively affect their institutional investments, particularly if they have significant experience in institutional investing.
5 sharesSource ↗
AI-powered roboadvisors are transforming wealth management by improving accessibility and efficiency, despite issues such as data privacy and regulatory obstacles.
4 sharesSource ↗
A new algorithm simplifies the calculation of sequence-space Jacobians in overlapping generations models, aiding in their analysis in general equilibrium.
4 sharesSource ↗
The Bank of Italy uses a combination of statistical models and expert assessments in its in-house credit assessment system to predict default probabilities of non-financial firms, aiding in monetary policy.
4 shares1 citation todaySource ↗
Large Language Models struggle with accuracy in specialized fields due to lack of specific knowledge in training data, a problem that can be solved by fine-tuning with domain-specific data.
3 shares1 citation todaySource ↗
A link between stochastic approximation and Kalman filtering has been found, leading to an online algorithm that adaptively tracks variances and achieves optimal learning rates.
4 sharesSource ↗
The impact of data analytics on audit quality depends on client consulting potential, with audit offices having high analytics capabilities more likely to disengage from high potential consulting clients.
3 sharesSource ↗
A new model combining quantum and classical machine learning has been proposed to improve the accuracy of credit default predictions in emerging markets.
2 sharesSource ↗
A comprehensive framework for Malliavin calculus has been developed, extending to infinite-dimensional Wiener space and enabling rigorous computational methods for stochastic analysis.
3 sharesSource ↗
A new method for decomposing residuals in linear regression models has been introduced, offering detailed analysis of prediction errors and an overfitting detection index.
3 sharesSource ↗
A study compares the role of AI and Machine Learning in hedge fund trading strategies, evaluating an AI-driven trading model against human-recommended trades.
2 sharesSource ↗
A new LogDet estimator has been proposed to address the challenges of handling high-dimensional samples in machine learning using entropy estimators.
2 sharesSource ↗
A novel theoretical framework uses quantum mechanics principles to model inflation and macroeconomic systems, treating economic goods and assets as quantum-like particles.
3 sharesSource ↗
A systematic review of the literature on the use of computer vision techniques in detecting victims in aviation accidents has been conducted, with a focus on search and rescue operations.
3 sharesSource ↗
New data and time series on global real estate returns over centuries have been presented, combining historical primary data with machine learning approaches and econometrics.
2 sharesSource ↗
The Farah Model is a new method for predicting price changes in financial markets, linking them to volatility and volume dynamics.
5 sharesSource ↗
A Bayesian calibration framework uses convolutional neural networks to efficiently quantify uncertainty and infer parameters in flood-prone areas with limited data.
5 sharesSource ↗
Young people are increasingly investing in mutual funds and stocks, but financial independence is difficult due to reliance on family support or personal savings.
4 sharesSource ↗
Monte Carlo Simulation is used to predict a company's financial outcomes for various events, with exotic derivatives offering new ways to reduce capital costs and increase shareholder return.
3 sharesSource ↗
The Ford Foundation and others plan to increase charity donations by issuing social bonds, a groundbreaking innovation in philanthropy.
4 sharesSource ↗
A course at The Chinese University of Hong Kong Shenzhen discusses the challenges and opportunities of factor investing in the age of big data and machine learning.
3 sharesSource ↗
A new approach for linear elasticity problems combines machine learning and the Matrix Discrete Empirical Interpolation Method to efficiently estimate problem output sensitivities.
3 sharesSource ↗
The study suggests that asset prices can be influenced by sentiment-driven preferences, which may explain inconsistencies in stock returns and economic fundamentals.
3 sharesSource ↗
The research presents a new method for detecting Android malware, using machine learning and permission analysis.
3 sharesSource ↗
The paper highlights the importance of robust data governance in banks for regulatory alignment, operational resilience, and accurate decision-making.
3 sharesSource ↗
The research shows that allowing the supply of financial assets to fluctuate over time can lead to persistent deviations in dividends.
3 sharesSource ↗
The study investigates the effectiveness of principal component analysis in cryptocurrency trading, suggesting potential for improvement.
3 sharesSource ↗
The paper discusses the use of longevity risk-sharing pools in retirement plans, emphasizing the need for methods to reduce payout volatility.
3 sharesSource ↗
The research explores the effect of the clamping function in cryptocurrency contracts, showing that no-arbitrage bounds persist even without transaction fees.
3 sharesSource ↗
The study assesses the use of machine learning in detecting accounting fraud, aiming to compare its effectiveness with traditional models.
3 sharesSource ↗
The study analyzes the relationship between petroleum prices and stock sector indices in Canada, Saudi Arabia, the US, and China, revealing diverse volatility interdependencies and fluctuating optimal portfolio weights and hedge ratios.
4 sharesSource ↗
The research uses traditional econometric models, machine learning, and deep learning techniques to predict financial time series, using SP 500 index and Bitcoin data, and assesses the models based on forecast error metrics and trading performance indicators.
3 sharesSource ↗
The paper explores the link between banks' ESG scores and their funding costs, concluding that higher ESG ratings positively influence funding costs and that changes in ESG ratings significantly impact banks' bond yields.
4 shares7 citations todaySource ↗
The study contrasts the traditional method of calculating logarithmic returns with the fractional differencing method in data preparation for machine learning models, finding that fractional differentiation methods enhance predictive model forecasting performance.
3 sharesSource ↗
The paper examines the economic functions of repo contracts and the structure of government bond repo markets, investigates the main factors influencing euro-area repo market dynamics, and discusses the role of central clearing services and new client clearing models in the repo market.
4 shares4 citations todaySource ↗
The research investigates the impact of temperature anomalies on firm-level risk, concluding that temperature beta affects firms' productivity, profitability, and cash flows, and that stocks with high temperature betas yield higher risk-adjusted returns.
3 sharesSource ↗
The article discusses the transformative effects of AI on the global financial system, including benefits and potential issues like algorithmic bias and data privacy.
3 sharesSource ↗
The research finds that dealers do not exploit private information in client-dealer relationships and such relationships do not necessarily improve client prices.
3 sharesSource ↗
The paper introduces two new methods for estimating stochastic volatility diffusions using Quantum-Inspired Classical Hidden Markov Models and Quantum Hidden Markov Models.
2 sharesSource ↗
The study suggests a new AMM protocol to eliminate liquidity providers' impermanent loss, after examining the profitability of automated market maker liquidity providers in ETH/USD pools.
2 sharesSource ↗
The article concludes that only a few could have succeeded in 401k and defined contribution plans without extreme risk tolerance and thrift.
2 sharesSource ↗
The research integrates risk and loss aversion into a model of household financial decision-making, finding that loss aversion significantly impacts participation, portfolio allocation, and savings decisions.
2 sharesSource ↗
Brexit Study: The paper indicates that exposure to financial markets can decrease support for populist policies, as shown by an RCT implemented before the Brexit referendum.
2 sharesSource ↗
The article presents a machine learning model for predicting biomass futures in salmon farming, aiding farmers in managing production risks.
4 sharesSource ↗
The research introduces an AI-enhanced shrinkage estimation for more efficient risk-optimized portfolio implementation, outperforming traditional methods.
4 sharesSource ↗
The study finds that the Minimum Variance Portfolio performs best in cryptocurrency selection, but lacks diversification.
3 sharesSource ↗
The paper calls for improved metrics to measure firms' impact on biodiversity, citing current methods' limitations and lack of understanding in business and finance.
3 sharesSource ↗
The article explores the paradox of Ukrainian banks' excessive activity in the government bonds market despite ample liquidity and positive financial results.
3 sharesSource ↗
The research analyzes firms' disclosure decisions and investors' incentives, concluding that the current equilibrium is socially inefficient.
3 sharesSource ↗
The paper proposes a new classification system for environmental scores, aiming to better assess unrated companies and guide investment strategies.
3 sharesSource ↗
Hilary Till discussed various risk management aspects at a meeting of the Professional Risk Managers International Association in Chicago.
2 sharesSource ↗
The paper proposes a structured asset pricing model that uses firm-level signals to estimate the stochastic discount factor and identify key pricing characteristics.
3 sharesSource ↗
The study finds that mutual funds often cross-trade in corporate bonds, which is beneficial during stressful times, but new regulations have reduced the associated cost savings.
2 sharesSource ↗
The paper identifies foreign investors' supply forces and domestic agents' demand forces as the main drivers of deviations in covered interest parity, based on micro-level transaction data.
3 sharesSource ↗
The study reveals that personal and professional traits of Chinese fund managers, such as gender, experience, and education, partially affect their performance and ability to generate excess returns.
3 sharesSource ↗
The proposed model suggests that changes in inventory and transaction costs can influence trading decisions, market stability, and indicators in over-the-counter markets.
2 sharesSource ↗
The study finds that during the Covid-19 period, Indian equity mutual funds increased their investments in foreign stocks, particularly US technology stocks, resulting in unusually high net returns.
2 sharesSource ↗
The study shows that family-owned firms have lower financing costs during periods of low market stress but face higher costs and are more likely to be downgraded during periods of high stress.
2 sharesSource ↗
Economics working papers from RePEc's NEP field reports.
30 items
A second-generation Automated Adaptive Trading System may stabilize emerging markets during downturns, countering challenges posed by algorithmic trading and passive investing.
27 sharesSource ↗
Machine learning has been utilized to pinpoint assets causing downward trends in the Pakistan Stock Exchange, suggesting an efficient asset allocation scheme.
25 sharesSource ↗
A novel risk parity portfolio optimization method using expected shortfall as the risk measure can lessen sensitivity to volatility shocks, decrease portfolio turnover during market turmoil, and enhance risk-adjusted returns.
16 sharesSource ↗
The research finds that trading strategies based on the Sharpe Ratio are more profitable than the buy-and-hold strategy in global markets, supporting the Adaptive Market Hypothesis.
15 sharesSource ↗
The study introduces a new method for assessing decision-making units' efficiency over time, using the Whale Optimization Algorithm to identify stable trading strategies and companies.
11 sharesSource ↗
The paper uses a new data augmentation technique to study poverty in the Middle East and North Africa, showing how alternative data sources can be used for poverty analysis when traditional income data is scarce or unavailable.
10 sharesSource ↗
The BRM method is introduced for analyzing incomplete data sets, showing improved predictive performance for both linear and nonlinear models.
20 sharesSource ↗
A new machine learning strategy, N-MDIS, is proposed for better equity premium prediction, outperforming existing MDIS and N-MDIS strategies.
19 sharesSource ↗
The study suggests that firms are more likely to adopt zero-leverage policies as product market competition increases, especially those with higher earnings volatility.
18 sharesSource ↗
The paper shows that both negative and positive news significantly impact intraday stock return volatility, reevaluating the role of news sentiment.
16 sharesSource ↗
A modified version of Stochastic Gradient Boosting is proposed for estimating production possibility sets in DEA, reducing overfitting and showing competitive performance against C2NLS.
16 sharesSource ↗
A study shows machine learning models are more successful than traditional methods in predicting Chinese corporate mergers and acquisitions.
28 sharesSource ↗
New deep learning frameworks have been proposed for better risk management in finance, outperforming existing methods in estimating Value at Risk and Expected Shortfall measures.
27 sharesSource ↗
Machine learning methods can accurately predict Chinese stock market volatility using the volatility of long-term treasury bond contracts, outperforming traditional models.
24 sharesSource ↗
The article proposes a new algorithm and machine learning model to improve the efficiency and accuracy of the Lot Streaming and Scheduling Problem (LSSP) with uncertain product arrival times.
16 sharesSource ↗
The paper introduces a new machine learning technique for analyzing and modeling complex time series, providing a potential alternative to the Box-Jenkins method, especially in financial modeling.
13 sharesSource ↗
The article uses machine learning to create a monetary policy frictions index from financial news, revealing that these frictions significantly impact the nonperforming loans of Chinese commercial banks.
12 sharesSource ↗
The paper uses machine learning to study the global housing market, finding that the US market is the main source of systematic shocks and its interest rate is the key global predictor of spillover intensities.
10 sharesSource ↗
ML vs. DL: Deep learning methods have been found to be more effective than traditional machine learning in predicting oil prices, especially during crises.
31 sharesSource ↗
The newly introduced MFF-CPPM model in China has shown higher accuracy and flexibility in predicting carbon trading prices compared to previous models.
10 sharesSource ↗
The article discusses a study that uses machine learning to predict the CBOE Volatility Index, highlighting the importance of weekly jobless claim data.
23 sharesSource ↗
The paper finds traditional machine learning models to be more effective than deep learning models in predicting Eurozone banking sector stock prices.
13 sharesSource ↗
The article suggests that AI capability directly affects firm performance, with a data-driven culture and AI infrastructure playing key roles.
5 sharesSource ↗
The research highlights the need for communication and a holistic approach to climate change, using machine learning to analyze social media discussions.
4 sharesSource ↗
The study investigates the use of dark patterns in retail investment, and how behavioral sciences and AI can improve regulation and investor protection.
2 sharesSource ↗
The research profiles young informal workers in the EU27, aiming to understand how Covid-19 has impacted youth employment informality.
2 sharesSource ↗
The article discusses how artificial intelligence can improve resource management in cloud environments, boosting DevOps workflows' performance and efficiency.
2 sharesSource ↗
The review investigates the link between e-governance initiatives and citizen participation, emphasizing the need for interdisciplinary research for effective evaluation.
2 sharesSource ↗
The paper analyzes factors affecting banks' performance, proposing new research areas in digital transformation, artificial intelligence, and the impact of COVID-19.
1 sharesSource ↗
The study tests the applicability of the Work Need Satisfaction Scale (WNSS) to online gig workers, suggesting the scale needs modification to suit online platform work.
1 sharesSource ↗
Repositories the letter featured.
10 items
The MCP server enables LLM agents to easily access and retrieve data from any database.
85 shares
A new tool uses Google's Gemini 2.5 Flash Preview model to convert complex PDF tables into clean HTML, preserving the original layout and data.
9 shares
The article provides 120 interactive Python coding interview challenges centered on algorithms and data structures, with additional Anki flashcards.
30,322 shares
LightlyTrain is a leading PyTorch framework that pretrains computer vision models on unlabeled data for industrial applications.
328 shares
A company analysis tool, powered by LangGraph and Tavily, uses a multiagent framework and Google's Gemini 2.0 Flash and OpenAI's GPT4.1 for detailed company research.
1,117 shares
Natural Language Web explores the incorporation of human language comprehension into web technology.
3,603 shares
CVPR 2025 论文和开源项目合集 compiles papers and open-source projects from the 2025 CVPR conference.
20,028 shares
An open protocol enabling communication and interoperability between opaque agentic applications investigates a new protocol for effective interaction between different software applications.
16,156 shares
A research prototype of a human-centered web agent introduces a prototype for a user-friendly web agent.
2,522 shares
LTXVideo Support for ComfyUI examines the integration of LTXVideo features into the ComfyUI interface.
1,961 shares
Industry news: funds, hiring, markets and regulation.
10 items
Citadel Securities saw a 45% increase in net trading revenue in Q1 2025, reaching $3.4bn due to ongoing market volatility.
7 shares
Cantor Fitzgerald plans to buy UBS's O'Connor alternatives investment platform, adding $11bn in various assets to its portfolio.
6 shares
Gavea Investimentos is changing its investment strategy to a more global approach due to increasing volatility.
6 shares
Temple Capital, a crypto-focused hedge fund, has hired high-profile staff from Brevan Howard, BlueCrest, and Hilbert Capital to cater to the growing interest in digital assets.
5 shares
Evolution Asset Management's flagship fund, Multi-Strategy No1, has seen a nearly 20% YTD increase and a total return of 1485% since 2015.
5 shares
The article discusses various cryptocurrency quantitative trading strategies and their influence on the blockchain market.
4 shares
The Shanghai Futures Exchange is considering allowing foreign hedge funds and institutional investors to access China's commodities markets to globalize the renminbi.
4 shares
Saba Capital Management has arranged a deal to offer a 100% tender at net asset value with CQS Natural Resources Growth & Income Trust following shareholder discussions.
3 shares
Due to concerns about fiscal imbalances and global supply dynamics, hedge funds and institutional investors are growing more bearish on US Treasury bonds.
3 shares
Impactive Capital intends to escalate its campaign at WEX Inc by proposing at least four directors to the board at the 2026 annual meeting.
3 shares
Episodes on markets, quant methods and economics.
10 items
Brad Barrie argues that true investment diversification is about diversifying return drivers, not just asset classes.
15 shares
Julian Brigden discusses U.S. exceptionalism, dollar dynamics, treasury market vulnerabilities, and global capital flows on a podcast.
10 shares
James Griffiths talks about the challenges and benefits of establishing an alternative data provider within a large corporation.
10 shares
Jonathan Shelon discusses the growth of covered call strategies and the potential of international markets, especially Chinese internet stocks.
10 shares
Cole Wilcox emphasizes the effectiveness of trend-following in stock investments and the importance of accepting losses to be a successful investor.
8 shares
Helen Thomas discusses political risk resurgence, S&P 500 derivative risks, and Trump's strategy in a podcast.
8 shares
Ross McDonald explains the application and future of money market funds in a podcast.
6 shares
Rob Jankiewicz discusses the creation of indexes and the parameters used in a podcast.
2 shares
Mike Green talks about the use and contradictions of tariffs in a MacroVoices podcast.
2 shares
Utkarshani Jaimini discusses the development of AI systems that understand cause and effect using knowledge graphs and neural networks.
1 shares