Quant Letter: June 2025, Week-3

Weekly (96th Edition)

SSRN

Recently Published

Quantitative

Bayesian VAR Forecasting: A new method for predicting and modeling time series of count data has been created, allowing for the simultaneous modeling of multiple variables and managing data irregularities. (2025-06-09, shares: 5.0)

Market Power in Electricity Markets: A study investigates the relationship between hedging and market power abuse in electricity markets, assessing the economic incentives to deviate from competitive behavior. (2025-06-09, shares: 4.0)

Liquidity Flows in Broker Dealers: Research indicates that internal borrowing rates within the same bank holding company are higher than external rates, implying that regulatory capital requirements increase the value of internal funding. (2025-06-09, shares: 3.0)

Electricity Price Volatility & Financial Stress: The research explores the link between electricity price changes and financial stress in Europe, suggesting market-based pricing and diverse energy supplies. (2025-06-09, shares: 2.0)

Firm Linkages: QCML vs. Euclidean Similarity: The first article presents a new technique, Characteristic Vector Linkages (CVLs), for estimating firm linkages, which when combined with Quantum Cognition Machine Learning (QCML), can create profitable trading strategies. (2025-06-09, shares: 3.0)

Recently Updated

Quantitative

Optimal Lotteries in Non-Convex Economies: A new method has been developed for solving optimal lotteries in models with nonconvexities, proving more efficient than traditional methods. (2025-04-17, shares: 36.0)

Quantum Machine Learning for Trading Strategies: The study investigates the use of quantum machine learning to optimize high-frequency trading strategies in US treasuries and forex markets. (2024-12-25, shares: 3.0)

Hedging Risks with Reinforcement Learning: A proposed deep reinforcement learning framework optimizes the hedging of specific risk factors in financial instruments using Shapley value decompositions. (2025-06-02, shares: 5.0)

Gamma Scalping for American Option Valuation: The paper highlights gamma scalping profitability as a crucial factor in the strategy and valuation for American-style options contracts. (2025-06-07, shares: 4.0)

Forecasting Count Data with Bayesian VAR: A new framework for modeling and forecasting time series of count data extends the traditional Vector Autoregression framework to accommodate count-like outcomes. (2025-06-09, shares: 5.0)

Big Data Analytics in Finance: The essay discusses the benefits of Big Data Analytics and predictive modeling in Risk Management for optimizing transactions in the banking sector. (2020-12-16, shares: 2.0)

Optimising Large Language Models: The article reviews the optimization strategies of Large Language Models, categorizing gradient-based and non-gradient-based methods and discussing future research. (2025-05-01, shares: 16.0)

Machine Learning in Market Crash Prediction: The article explores how Machine Learning can be used to predict market crashes, detailing the complexities and techniques involved. (2024-06-01, shares: 2.0)

Quantum Model for Credit Default Prediction: The study suggests a hybrid quantum-classical machine learning model to improve the accuracy of credit default predictions in emerging markets. (2025-05-28, shares: 2.0)

Kelly Betting with Recovery Constraints: The paper suggests a modified Kelly optimization that balances long-term growth with short-term recovery risk in skewed return environments. (2025-06-03, shares: 6.0)

Reference-dependent Preferences in Asset Pricing: The article presents a model that explains how reference-dependent preferences can lead to sentiment-driven asset prices, solving several empirical puzzles in asset pricing. (2023-02-01, shares: 3.0)

The Hype Index for Market News Attention: The paper introduces the Hype Index, a metric that uses Natural Language Processing to measure media attention towards large-cap equities and extract predictive signals from financial news. (2025-05-30, shares: 3.0)

Financial

HighDimensional Finance Learning: The article investigates the use of machine learning for financial forecasting, focusing on the role of within-sample standardization in Random Fourier Features and the analysis of ridgeless regressions. (2025-06-04, shares: 8.0)

Volatility Spillovers Modelling: The research compares four GARCH methods in modeling the relationship between petroleum prices and stock indices in Canada, Saudi Arabia, the US, and China, highlighting diverse volatility interdependencies. (2025-05-29, shares: 4.0)

RiskFree Asset-Less Portfolio Theory: The paper introduces a new portfolio theory that considers the lack of a universally accepted risk-free asset, suggesting safety is an investor-specific property that varies across different boundaries. (2025-06-07, shares: 3.0)

Technological Usefulness Learning: The study uses natural language processing and machine learning to create a technology dataset from patent descriptions and U.S. public firms, uncovering the core technologies of non-patenting firms. (2025-06-07, shares: 5.0)

RealTime Option IV Surface Modeling: The article proposes a two-step forecasting framework for the option implied volatility surface, which can handle large datasets and high data frequencies, and performs better than random walk forecasts. (2025-05-26, shares: 3.0)

Dynamic Currency Arbitrage: The research identifies widespread mispricing in currency markets using a conditional latent factor model, showing that currency characteristics contribute more to mispricing than macroeconomic fundamentals. (2025-03-01, shares: 3.0)

Commodity Futures Investment: Hilary Till discusses the commodity investment universe, investment focus, return rationale, investment process, return composition, portfolio construction, and risk management at the Alternative Investments Group of Calyon Financial. (2004-09-02, shares: 3.0)

Deep IV Factor Models: The article proposes a model using neural networks and linear regression to better estimate daily volatility of stock options, especially during earnings announcements and sparse data periods. (2025-06-05, shares: 2.0)

Portfolio Moments Relaxation: The study presents a method to approximate portfolio skewness and other higher odd moments, demonstrating how incorporating skewness can increase the optimal portfolio's skewness. (2025-06-06, shares: 2.0)

ArXiv

Finance

Aggregated Snell Envelopes: The article discusses the creation of an aggregator for Snell envelopes in a non-dominated setting, used to establish a reliable hedging strategy for American-style options in a semi-martingale setting. (2025-06-17, shares: 11)

Optimal Execution: The research investigates the best strategy for buying a large number of shares over a set time, considering factors like price impact and market conditions, and uses numerical examples to demonstrate the findings. (2025-06-13, shares: 9)

Implied Probabilities in Credit Risk: The paper outlines a two-stage method for pricing credit risk using the Merton model, introducing a new mapping between risk-neutral and physical parameters for stress testing and credit risk analysis. (2025-06-15, shares: 8)

Choquet Rating and Risk Consistency: The study explores the concept of risk consistency in Choquet rating criteria, providing a comprehensive analysis of Choquet risk measures and rating criteria that meet risk consistency standards. (2025-06-16, shares: 7)

Quantum BSDE Solver for High-Dimensional PDEs: The research introduces a quantum machine learning method for approximating solutions to complex partial differential equations, showing that Variational Quantum Circuits offer better accuracy and lower variance, especially in highly nonlinear situations. (2025-06-17, shares: 6)

Small Volatility Approximation in Multi-Factor HJM Models: The paper showcases the use of Small Volatility Approximation in calibrating the Multi-Factor HJM model, highlighting that the calibration quality is high and independent of the number of factors. (2025-06-14, shares: 5)

Credit Risk for Green and Brown Loan Portfolios: The study presents a credit risk model for portfolios of green and brown loans, expanding the ASRF framework and demonstrating how value-at-risk is influenced by various factors, providing a foundation for future credit risk modeling advancements. (2025-06-14, shares: 5)

Economics

Network Experimentation Methods: The article explores different techniques for conducting experiments that involve interaction networks between subjects. (2025-06-12, shares: 22)

Dynamic Allocation Model: The article introduces a complex model for forecasting the potential risks in the global financial market, enhancing global asset allocation strategies. (2025-06-14, shares: 13)

High SES Consumption: The article presents empirical data showing a strong correlation between higher income and diverse consumption habits across various brands and price points. (2025-06-16, shares: 10)

EconGym AI Testbed: The article presents EconGym, a scalable testbed that integrates various economic tasks with AI algorithms for large-scale simulations and policy optimization in economic research. (2025-06-13, shares: 8)

Predicting Inflation: A study shows machine learning algorithms, particularly the Extreme Gradient Boosting model, are more effective than traditional methods in predicting Indonesia's inflation. (2025-06-12, shares: 8)

Price Adjustment: Research indicates that small price changes asymmetry varies with the business cycle, with more asymmetry during low unemployment periods, implying firms' pricing behavior is influenced by the economy. (2025-06-12, shares: 8)

The U.S. Phillips Curve: A study using MSA-level panel data investigates if the U.S. Phillips Curve's slope changed during and post-COVID-19, providing insights into the unemployment cost of disinflation. (2025-06-16, shares: 7)

Incentivizing Flexibility: Research examines the impact of a centralized capacity market and an advanced reliability reserve on investments in demand-side flexibility technologies in the power sector, favoring the advanced reliability reserve as a more effective solution. (2025-06-17, shares: 6)

AI for FDI Facilitation: The article suggests an AI system that uses OCR and Large Language Models to simplify the verification of tariff exemptions for Foreign Direct Investment in manufacturing, enhancing operational efficiency. (2025-06-12, shares: 12)

Dynamic Reinsurance Treaty: The paper introduces a multi-agent reinforcement learning framework for reinsurance treaty bidding, showing its ability to enhance risk transfer efficiency and surpass traditional pricing methods in reinsurance markets. (2025-06-16, shares: 6)

Crypto & Blockchain

Cryptocurrency Options Pricing Models: Research indicates that Kou and Bates models, which include jumps and stochastic volatility, are the most accurate for pricing Bitcoin and Ether cryptocurrency options. (2025-06-17, shares: 6)

DeFi Protocol Risk Management: The latest versions (v3) of Aave and Compound lending protocols show improved risk management compared to their previous versions (v2), with liquidation events boosting total value and revenue, particularly on the L2 blockchain. (2025-06-15, shares: 5)

Historical Trending

Recession Detection with Classifiers: The article introduces a novel method for real-time detection of US recessions using unemployment and vacancy data, predicting a 71% chance of a current recession based on May 2025 data. (2025-06-11, shares: 25)

TrendFolios Framework for Portfolios: The study presents a portfolio construction framework using momentum and trend-following signals across various asset classes, showing its potential to generate excess returns and manage risk over 22 years. (2025-06-11, shares: 19)

Interest Rate Announcements Communication Quality: The paper analyzes the Bank of Israel's interest rate announcements using text-mining techniques, finding them more comprehensible than those of the Federal Reserve and European Central Bank, and their sentiment aligns with economic fluctuations. (2025-06-11, shares: 19)

Nonconvex Game for Ancillary Markets: The study uses noncooperative game theory to characterize zonal ancillary market coupling, finding that multi-agent deep reinforcement learning leads to lower market costs but higher profit allocation variability. (2025-06-11, shares: 18)

ArXiv ML

Historical Trending

Rankify: The article introduces Rankify, an open-source toolkit designed to integrate retrieval, re-ranking, and RAG processes, aiming to improve retrieval and re-ranking methodologies while ensuring consistency and ease of use. (2025-02-04, shares: 21)

Decision Theory: The article develops a connection between uncertainty quantification using prediction sets and risk-averse decision-making, introducing an algorithm, Risk-Averse Calibration (RAC), to optimize action policies from predictions within a user-defined risk limit. (2025-02-04, shares: 20)

QGuided Stepwise Search for Language Agents: QLASS system enhances the performance and efficiency of language agents by using Q-values to provide step-by-step guidance, even with limited supervision. (2025-02-04, shares: 188)

Reliability of Large Language Model Benchmarks: The study introduces platinum benchmarks, designed to reduce label errors and ambiguity, to improve the accuracy of large language model assessments. (2025-02-05, shares: 55)

Masked Autoencoders for Diffusion Models: The MAETok system uses an autoencoder to create a semantically rich latent space, enhancing the quality of high-resolution image synthesis. (2025-02-05, shares: 38)

Algebraically Converging Stochastic Gradient Descent: A new gradient descent algorithm with adaptive randomness tuning enhances the global convergence rate for nonconvex optimization problems. (2022-04-12, shares: 28)

BFSProver: The article presents BFS-Prover, a framework for automatic theorem proving using Lean4, showing that Best-First Tree Search can perform well in large-scale theorem proving tasks. (2025-02-05, shares: 22)

GitHub

Finance

Pixeltable Data Infrastructure: Pixeltable is an AI system that uses a step-by-step method to manage various types of workloads. (2023-05-10, shares: 386.0)

Alchemist Trading System: Alchemist is a fast, automated trading system that uses Ray technology. (2025-02-26, shares: 7.0)

Proactive Time Series Forecasting: KDD25 is a new system for predicting online time series that can adapt to changes in concepts. (2024-12-12, shares: 24.0)

AutoHedge: The article explores automated methods for risk protection and performance testing in financial trading. (2019-09-24, shares: 64.0)

OCR Layout Analysis: The article introduces an OCR tool that can identify reading order tables in 90 languages. (2024-01-10, shares: 17612.0)

Trending

Amazing OSINT List: The article presents a detailed list of exceptional Open Source Intelligence tools and resources. (2016-11-30, shares: 21656.0)

Node: The article delves into the functionalities and features of the Node.js JavaScript runtime. (2014-11-26, shares: 111698.0)

Whisper: The article outlines the procedure of converting OpenAI's Whisper model into C language. (2022-09-25, shares: 40703.0)

Coding Environments: The article investigates coding environments for multiple, independent, and secure agent operations. (2025-05-23, shares: 1600.0)

Python PGlite Wrapper: The article presents a PGLite wrapper in Python for lightweight app testing with Postgres, similar to SQLite. (2025-06-05, shares: 420.0)

Twitter

Quantitative

Optimal Kelly Leverage for SP 500: The best Kelly leverage for daily SP 500 returns between 1997 and 2024 is around 2.4, as higher values decrease long-term growth due to increased volatility and drawdowns. (2025-06-17, shares: 3)

Lowvol Factor in Asset Pricing: A new study emphasizes the importance of the low-volatility factor in asset pricing models, especially in relation to factor asymmetry and frictions. (2025-06-16, shares: 1)

Miscellaneous

Trend Following and Drawdowns: ManGroup analyzes the current situation of trend following and drawdowns, questioning if the current scenario is unique. (2025-06-17, shares: 1)

SEAL: LLMs Updating Weights: SEAL, a new framework, enables LLMs to create their own training data and adjust their weights based on new inputs, using the improved model's performance as a reward. (2025-06-13, shares: 1)

Investing Research Roundup: The recent investment research roundup discusses topics like predicting cryptocurrency using sentiment, a strategy based on foreign exchange mispricings, multiple option-based predictors, a regime-switching model, and more. (2025-06-17, shares: 0)

Reddit

Quantitative

Risk Management (2025-06-14, shares: 121.0)

Hedge Fund Strategies (2025-06-11, shares: 128.0)

Tower Research Opportunity (2025-06-12, shares: 61.0)

Jane Street Manipulation (2025-06-15, shares: 218.0)

PineScript Framework Feedback Request (2025-06-11, shares: 47.0)

Rising

Quant Finance Acquisition (2025-06-13, shares: 113.0)

Quant Job Opportunities (2025-06-11, shares: 98.0)

XTX Markets vs. Optiver (2025-06-14, shares: 117.0)

Resume Critique (2025-06-14, shares: 104.0)

Accuracy Trends (2025-06-12, shares: 91.0)

Paper with Code

Trending

InfLLM v2: Longcontext Processing: The article introduces InfLLM v2, a new model with a trainable sparse attention mechanism designed for quicker processing of long-context data. (2025-06-12, shares: 7887.0)

TradingAgents: Financial Trading Framework: The second article explores the progress in automated problem-solving using societies of agents driven by large language models (LLMs). (2025-06-12, shares: 3721.0)

RWKV Goose: State Evolution: The third article presents RWKV7 Goose, a new sequence modeling architecture that ensures consistent memory usage and inference time per token. (2025-06-12, shares: 2647.0)

TabM: Tabular DL Advancement: The article explores different deep learning structures for managing and learning from structured data, including basic and advanced models like Transformers. (2025-06-14, shares: 384.0)

Rising

PixelsDB Data Analytics: The article explores a serverless query engine that performs queries and provides diverse pricing based on performance service levels. (2025-06-14, shares: 213.0)

PreTraining Framework for Agentic Search: The Retrieval Augmented Mask Prediction (RAMP) task improves Large Language Models' retrieval and reasoning skills by teaching them to use search tools during the pretraining stage. (2025-06-12, shares: 108.0)

Spiking Graph Convolution Networks: Graph Convolutional Networks (GCNs) excel due to their superior capacity to learn and process graph information. (2025-06-14, shares: 103.0)

Autonomous Knowledge Graph Construction: AutoSchemaKG is a novel framework that enables the autonomous creation of knowledge graphs, removing the requirement for predefined schemas. (2025-06-17, shares: 69.0)