Quant Letter
The week in quantitative finance research: 90 papers from arXiv, SSRN and RePEc, screened and ranked, 26 Sep – 2 Oct 2026.
This week centers on AI's expanding role in trading and markets, alongside persistent microstructure patterns and monetary policy transmission. Key papers: "AI Trading Methods: Backtests Versus Real Markets" measures the live-market gap for machine learning and LLM strategies; "Certified Alpha Capacity and Decay" derives exact deployment thresholds before signal decay; "AlphaPareto" applies reinforcement learning to multi-objective alpha discovery. Corporate bond signals predict equity returns; FOMC surprises persist; US rate shocks widen global spreads.
If you read five papers this week, read these.
The research measures when a trading signal accumulates enough statistical evidence for deployment before its economic value decays, deriving exact feasibility thresholds.
Signals extracted from corporate bond portfolios predict next-month equity returns of same issuers at 21 basis points higher after controlling for stock characteristics.
Topics drawing unusually many papers this week.
0.9% of this week's 774 new papers, up from almost none over the previous week.
3.6% of this week's 774 new papers, 2.1x its share over the previous week (1.7%).
4.1% of this week's 774 new papers, 1.8x its share over the previous week (2.2%).
3.6% of this week's 774 new papers, 1.8x its share over the previous week (2.0%).
5.7% of this week's 774 new papers, 1.5x its share over the previous week (3.8%).
What the quant community starred on GitHub this week, from a panel of 10,589 developers who follow financial machine learning (244 of the 860 we checked starred something). Counts only: we never publish who starred what.
Collection of 5800+ trading strategies.
★ 335+335 this weekNew repo
Open-source quantitative database framework for China stocks.
★ 269+26 this weekPythonNew repo
Grid search and backtesting for moving-average crossover strategies.
★ 7+4 this weekPythonNew repo
AI trade decision agent on Monad blockchain.
★ 2,748+325 this weekTypeScriptNew repo
Fly connectome simulation with Coinbase agent trading.
★ 865+39 this weekPythonNew repo
Cross-provider market data validation for OHLCV prices.
★ 16+16 this weekPythonNew repo
Self-hosted A-stock selection monitoring and backtesting platform.
★ 5,365+182 this weekPython
Agent memory system that learns from interactions.
★ 44.5k+14.5k this weekPython
No description available.
★ 2,719+1,395 this weekPythonNew repo
Safe private runtime for autonomous AI agents.
★ 14.3k+5,482 this weekRust
Non-autoregressive decision engine for text classification over 100 languages.
★ 30.1k+5,194 this weekPythonNew repo
Foundation models for structured data.
★ 270+246 this weekPython
CLI for coding agents to find code by description.
★ 2,041+2,040 this weekTypeScriptNew repo
Now published in Journal of Industrial and Management Optimization · featured 9 Oct 2025, 2 days after release; 1 citations
Now published in Applied Energy · featured 13 Sep 2025, 5 days after release
Now published in BMJ Public Health · featured 12 Nov 2025, 5 days after release
Filter by venue, topic or fanfare.
No papers match these filters.
Quantitative finance and ML-for-finance preprints
30 papers · 30 figures
The research measures when a trading signal accumulates enough statistical evidence for deployment before its economic value decays, deriving exact feasibility thresholds.
Fanfare 4Trading, Microstructure & ExecutionNicolò BonacorsiPDF
A survey traces how market design has evolved from floor trading through electronic exchanges to AI agents, examining whether markets can allocate goods efficiently without full revelation.
Fanfare 3Trading, Microstructure & ExecutionIrene AldridgePDF
The research shows when to follow or fade financial narratives by studying the covariance between institutional statements and revealed trading positions using machine learning.
Fanfare 3Trading, Microstructure & ExecutionAli Atiah AlzahraniPDF
A lightweight module adds confidence estimates to limit-order-book forecasters, raising directional F1 by 0.11-0.15 when predicting the most confident 10% of trades.
Fanfare 3Trading, Microstructure & ExecutionDerrick Gilchrist Edward Manoharan et al.PDF
A method generates realistic limit order book messages guaranteed to be consistent with current market state, achieving 100% replayability and 2.7–3.6× speedup over existing approaches.
Fanfare 3Trading, Microstructure & ExecutionJunoh Kang et al.PDF
Measurement of over a year of CME market data reveals transactions cluster within microseconds, yielding design principles: single-thread receivers never queue, two-thread splits reduce tail latency.
Fanfare 3Trading, Microstructure & ExecutionVincent MaciejewskiPDF
Reinforcement learning agent combining policy optimization with regime priors controls drawdown while blending learned and rule-based portfolio exposure on held-out equity data.
Fanfare 3Trading, Microstructure & ExecutionDuong Hien Chi Kien and Thanh Trung HuynhPDF
The study shows that reinforcement-learning traders in limit order books create phase transitions between orderly trading and volatile cascades, with non-linear market impact dynamics.
Fanfare 2Trading, Microstructure & ExecutionJan RosenzweigPDF
A benchmark compares machine learning, reinforcement learning, and large language model trading methods across historical backtests, paper trading, and live markets to measure the gap.
Fanfare 4ML & AI MethodsXingtong Yu et al.PDF
The research uses reinforcement learning with multi-objective rewards to discover formulaic alphas that work well together despite evolving reward functions and shifting environments.
Fanfare 3ML & AI MethodsYingbo Zhao et al.PDF
A new method adapts financial language models to avoid look-ahead bias using low-rank adapters instead of expensive annual retraining, matching performance of full pretraining.
Fanfare 3ML & AI MethodsSeunghan Lee et al.PDF
Cross-model rank correlations for LLM-extracted sentiment, clarity and risk average only 0.52, and model choice significantly alters coefficient signs and significance in downstream analysis.
Fanfare 3ML & AI MethodsHamid Boustanifar and Sasan MansouriPDF
Diffusion transformer foundation model reformulates candlestick prediction as conditional trajectory generation, achieving 0.057 mean return RankIC across markets and timescales.
Fanfare 3ML & AI MethodsBoyu Zhang and Haorui LiPDF
The study tests whether extra reasoning in large language models improves portfolio returns net of trading costs across multiple model families, finding no reliable gains.
Fanfare 3ML & AI MethodsJiayi Chen and Guiling WangPDF
An evaluation framework tests language model agents on structured option trading tasks, finding current systems underperform in most real-world scenarios.
Fanfare 3ML & AI MethodsHaochen Luo et al.PDF
A framework tests whether self-evolving research capabilities improve alpha discovery over time, finding no consistent gains from accumulated experience and tools.
Fanfare 2ML & AI MethodsSiyuan Li et al.PDF
A hybrid approach combining neural networks with kernel methods produces robust derivative hedges in low-data regimes and scales via random Fourier features with convergence guarantees.
Fanfare 2ML & AI MethodsJean-Loup Dupret et al.PDF
Post-training Qwen3-4B via supervised fine-tuning and policy optimization more than doubles direction-magnitude score on chronological stock-price predictions to 43.31.
Fanfare 2ML & AI MethodsJiacheng Guo et al.PDF
Training teaches a language model to improve event forecasting by learning which evidence to gather, beating frontier models on hard questions at lower cost.
Fanfare 3Econometrics & ForecastingYusuf Afifi et al.PDF
Dual-path language model extended with financial tokens and news-conditioned revision achieves lowest error in 8 of 12 equity and energy forecasting settings across multiple horizons.
Fanfare 2Econometrics & ForecastingWentao Zhao et al.PDF
The research makes conditioning information a decision variable subject to constraints, solving how look-ahead bias and information availability affect portfolio choice and causal identification.
Fanfare 3Portfolio & AllocationAlejandro Rodriguez DominguezPDF
The study develops a Markov decision process framework for dynamic bond portfolio management that balances yield and interest-rate risk subject to transaction costs.
Fanfare 2Portfolio & AllocationBalaji Ramachandran et al.PDF
A deep learning framework for options trading enforces portfolio-level delta neutrality and other risk constraints during training, improving risk-adjusted returns with lower directional exposure.
Fanfare 2Portfolio & AllocationWee Ling Tan et al.PDF
Kernel scenario analysis estimates quantile predictions for portfolio stress-test gains, with online calibration to capture dependence between stressed and unstressed risk factors.
Fanfare 2Portfolio & AllocationZhongze Cai et al.PDF
A feedback model explains extreme commodity futures prices like negative oil by connecting delivery constraints to roll options and long-short position imbalances.
Fanfare 3Derivatives & VolatilityIosif Zimbidis and Ronnie SircarPDF
The study shows how hedging a derivative via trading the underlying changes its payoff when execution generates price impact and costs, deriving pricing equations and replication conditions.
Fanfare 2Derivatives & VolatilityDavid Itkin and Leandro Sánchez-BetancourtPDF
Proprietary AMMs earn 0.37 to 1.19 basis points within two seconds by repricing continuously and avoiding arbitrage, compared to passive AMMs losing 0.22 to 0.62 basis points.
Fanfare 3Crypto & DeFiOzan Solmaz et al.PDF
Analysis of Uniswap pools shows passive liquidity providers underperform aggregate pool returns, with the gap wider on Ethereum than on layer-two blockchains.
Fanfare 3Crypto & DeFiAgathe Sadeghi et al.PDF
A framework shows when oracle-informed automated market makers reduce loss versus rebalancing and improve capital efficiency, with counterfactual backtests on SPY data.
Fanfare 3Crypto & DeFiHamed Amini and Zachary FeinsteinPDF
Large language models outperform traditional methods at extracting missing financial data from SEC filings, with accuracy improving as model size matches document complexity.
Fanfare 3LLMs & TextPrisha Nair and Roee ShragaPDF
New working papers in finance, economics and ML
30 papers
Signals extracted from corporate bond portfolios predict next-month equity returns of same issuers at 21 basis points higher after controlling for stock characteristics.
Fanfare 4Asset Pricing & FactorsJun Kyung Auh and Woojung Kim
The study exploits Ant Group's suspended IPO in November 2020 as a natural experiment, finding that highly exposed firms suffered roughly 21 percentage point abnormal returns and experienced a 42% contraction in shadow-loan balances.
Fanfare 3Asset Pricing & FactorsYang Lio
A lognormal model shows that individual stocks carry an idiosyncratic variance penalty making most underperform the index, with about one-third beating the market over five years.
Fanfare 3Asset Pricing & FactorsGabriele Susinno
A reproducible pipeline for factor research finds no model reliably beats simple approaches after accounting for transaction costs in point-in-time factor tests.
Fanfare 2Asset Pricing & FactorsAmit Kumar Dudi
Filing discipline characteristics built from SEC timestamp data alone predict cross-sectional stock returns with a net Sharpe ratio of 1.27, concentrated in smaller and more volatile stocks.
Fanfare 2Asset Pricing & FactorsLyuge Zhu
Order flow imbalance predicts contemporaneous mid-price changes in Kalshi binary event contracts, with explanatory power varying from 0.29 for sports to 0.02 for macroeconomic events.
Fanfare 3Trading, Microstructure & ExecutionAsh Malék
Correcting for timestamp mismatch between ETF and Bitcoin markets, the study finds that same-day flows predict next-day returns at 1.67 percentage points per billion of inflow, while prior Bitcoin returns predict flows.
Fanfare 3Trading, Microstructure & ExecutionJuraj Fabus et al.
Bitcoin and Ether trading activity on major venues increases during the UTC window aligned with New York market open, with the effect appearing on both assets.
Fanfare 3Trading, Microstructure & ExecutionYuntao Wang
A survey traces how market design rules evolved from floor trading to limit order books to automated market makers to agent-based trading systems.
Fanfare 2Trading, Microstructure & ExecutionIrene Aldridge
On-chain prediction market analysis finds 68.2 percent of volume sits in well-calibrated markets with average seven-day pricing error of 0.03.
Fanfare 2Trading, Microstructure & ExecutionDavide Mancino et al.
Multilevel order flow imbalance in Indian equity derivatives predicts mean-reverting price dynamics with information coefficients growing from -0.004 at ten seconds to -0.032 at sixty seconds.
Fanfare 2Trading, Microstructure & ExecutionAbhimanyu Chaudhari
The study models how capital requirements tax banks on tailored loans, pushing riskier firms toward private credit; evidence shows that tighter leverage rules reduce bank tailoring by 25 percent in quantitative terms.
Fanfare 4Risk, Credit & BankingFrancesco Beraldi et al.
The research develops a four-layer diagnostic framework combining network topology, dynamic causality, tail risk, and regime classification to monitor multi-asset systemic stress in real time.
Fanfare 2Risk, Credit & BankingYouness Yachruti
Calibrated simulations show existing central bank digital currencies remain below thresholds needed to measurably affect bank credit or financial stability.
Fanfare 2Risk, Credit & BankingArmand Yawo Isaac Zewu
Analyzing 261 FOMC announcements shows that immediate stock and yield impacts from policy surprises persist, while post-announcement drift and monetary momentum unwind within days or weeks.
Fanfare 3Macro-Finance & RatesArka Prava Bandyopadhyay
The study shows that US federal funds rate shocks widen bid-ask spreads on equities across 37 markets for up to two months, while longer-maturity yield surprises reprice equities without affecting liquidity.
Fanfare 3Macro-Finance & RatesJames Brugler et al.
US monetary tightening triggers larger currency depreciation and sovereign stress in emerging markets when global public debt is high and foreign-currency debt exposure is elevated.
Fanfare 3Macro-Finance & RatesOkan Akarsu and Hatice Karahan
The research decomposes bond returns into real rates, risk premia, and inflation expectations, finding that forward guidance and asset purchases had opposite effects at the zero lower bound versus normal times.
Fanfare 3Macro-Finance & RatesJialing Wang et al.
Rollover clock measurement of consolidated Treasury and central bank liability repricing predicts U.S. Treasury interest rates and prices inflation costs of fiscal deficits.
Fanfare 3Macro-Finance & RatesNingpei Ding
Global connectedness across energy, commodity, carbon, and equity markets averages 52.4 percent and spikes to approximately 95 percent during major geopolitical disruptions.
Fanfare 2Macro-Finance & RatesMengxi GAO
The research finds that the first five trading days of each month account for 73% of individual bond credit returns and 83% of the market credit premium, revealing a concentrated timing pattern in fixed-income compensation.
Fanfare 3Derivatives & VolatilityAlexander Dickerson and Yoshio Nozawa
Using Kalshi FOMC contract probabilities, the paper shows that cross-outcome variance in Federal Reserve policy expectations contains significant information about long-run stock market volatility.
Fanfare 3Derivatives & VolatilityFuli Yang
Static out-of-the-money put options provide crash protection but drag long-term returns, while dynamic tail-risk strategies adapt to market regimes and deliver superior risk-adjusted returns across cycles.
Fanfare 2Derivatives & VolatilityHenry Mayambu
Deep residual networks outperform shallow models in asset pricing, achieving a long-short Sharpe ratio of 2.07 versus 1.92 for shallow versions by preserving and refining earlier layers.
Fanfare 3ML & AI MethodsDexin Peng and Xiaoyu Wang
The research proposes a temporal graph neural network with explainability tools for real-time systemic risk surveillance, achieving early warning signals 3-4 quarters ahead of financial distress on bank data.
Fanfare 3ML & AI MethodsGnana Praveena Nethala et al.
Computing capacity for 35 anomalies reveals that high-alpha long legs support less than $2 billion before trading costs eliminate excess returns, while lower-alpha strategies support billions more.
Fanfare 3Portfolio & AllocationDoron Avramov et al.
Investors using machine-learning forecasts can achieve Sharpe ratios of 1.2 by adjusting the number of portfolio groups based on the forecast's information coefficient, beating standard decile sorts.
Fanfare 2Portfolio & AllocationLukas Salcher et al.
A new 234-year synthetic total-return dataset for equities and Treasuries reveals that permanent 3× leverage faces near-total loss in most reconstructions, ranking below 2× on risk-adjusted returns.
Fanfare 2Econometrics & ForecastingChemicalStats Von der Mauerstrasse
The research develops a compact Bayesian VAR for real-time forecasting of GDP growth, inflation, unemployment, and the federal funds rate, with evaluation emphasizing density scores and predictive-interval diagnostics.
Fanfare 2Econometrics & ForecastingCenk Ufuk Yildiran
The research shows that adjusted illiquidity—price impact scaled by market absorption capacity—predicts which cryptocurrency tokens survive better than volatility, improving out-of-sample forecasting of token death.
Fanfare 3Crypto & DeFiChang Liu
Working papers curated by RePEc's NEP field reports
30 papers · 21 figures
Using ChatGPT's launch as a natural experiment, the paper finds that hyperscalers saw borrowing costs decline while software firms faced worse terms as debt markets repriced AI winners and losers.
Fanfare 4Asset Pricing & FactorsMarco Albori and Andrea ZaghiniPDF
The research shows that firms' hurdle rates exceed their financial cost of capital due to innovation risk and imperfect pledgeability, explaining weak productivity growth and declining business dynamism.
Fanfare 2Asset Pricing & FactorsCraig A. Chikis et al.PDF
AI-intensive patents generate delayed productivity and employment gains alongside falling consumer prices, with substantially larger aggregate effects than broader ICT shocks but reducing labor share and increasing wealth inequality.
Fanfare 4Econometrics & ForecastingAndrea Gazzani and Filippo NatoliPDF
The study shows that increased Federal Reserve communication lowers monetary policy uncertainty and generates substantial real effects: industrial production rises 0.3 percent and unemployment falls 0.2 percentage points within two months.
Fanfare 3Econometrics & ForecastingDogukan GuneyPDF
Analysis of a large macro-financial dataset ranks non-linear monetary transmission channels, finding transmission to long-term rates weakens at high interest rates and high credit growth, with sovereign risk mattering in the euro area.
Fanfare 2Econometrics & ForecastingDilan Aydin Yakut et al.PDF
The GARCH-MIDAS model incorporating a quarterly news-based Price Conflict Index outperforms benchmarks for forecasting US stock volatility over 150 years of monthly and daily data.
Fanfare 2Econometrics & ForecastingAfees A. Salisu et al.PDF
A pre-trained time-series model reading central bank published paths cuts forecast errors better than the banks themselves and hard-conditioned VARs, revealing exploitable institutional differences.
Fanfare 2Econometrics & ForecastingVegard H. Larsen and Leif Anders ThorsrudPDF
The paper shows that monetary policy reduces perceived tail risk for bank equity, encouraging banks to originate riskier loans to commercial and industrial borrowers.
Fanfare 3Risk, Credit & BankingXudong An et al.PDF
The study shows that banks with high foreign exchange risk reduce lending to both exposed and unexposed firms after exchange rate shocks, with measurable real effects on small and medium enterprises.
Fanfare 3Risk, Credit & BankingBurak Deniz et al.PDF
UK firm-level analysis shows that interest rate spreads are less sensitive to capital-to-debt ratios for firms with higher intangible intensity, suggesting intangibles are less effective collateral than tangible assets.
Fanfare 2Risk, Credit & BankingSara Holttinen et al.PDF
The research distinguishes solvency risk from systemic risk in European life insurers, finding growing systemic risk exposure since 2007 and evidence of interconnectedness with banks that intensifies during financial stress.
Fanfare 2Risk, Credit & BankingSomnath Chatterjee and David HumphryPDF
A theoretical model shows that debt accumulation during booms delays post-crash recovery through debt overhang and coordination failures, with debt restructuring conditional on recapitalization being more efficient than unconditional subsidies.
Fanfare 2Risk, Credit & BankingKeiichiro Kobayashi and Tomoyuki nakajimaPDF
Post-GFC, banks facing stricter regulation increased cov-lite loan issuances due to liquidity advantages that lower credit spreads, particularly for private firms seeking easier asset sales.
Fanfare 2Risk, Credit & BankingRobert Prilmeier and René M. StulzPDF
Integrating financial news embeddings from large language models with market data, the study improves systemic risk forecasts using conditional value-at-risk without requiring large datasets.
Fanfare 3LLMs & TextJunyu Chen et al.PDF
Large language models provide nearly identical portfolio advice across twenty-one countries despite local differences, following retail finance conventions rather than academic prescriptions and ignoring household balance sheets.
Fanfare 2LLMs & TextClaes BäckmanPDF
Large Language Models extract signals about potential trigger events from financial news, improving forward-looking estimates of downside risks and helping monitor financial stability threats ahead of major events.
Fanfare 3ML & AI MethodsDomenic Kellner et al.PDF
AI automation reduces occupational employment by 21 percent with little wage effect, while AI augmentation raises wages by 8 percent, revealing that AI's labor impact depends on the balance between these opposing channels.
Fanfare 3ML & AI MethodsTomáš Oleš and Francesco RonconePDF
Testing eight volatility models on equity indices with prespecified holdout periods, the research finds that gains from more complex models often do not persist across markets or time.
Fanfare 3Derivatives & VolatilityHonfei Guo et al.PDF
The research shows that past asset price trajectories predict implied volatility movements up to two years forward, with a parsimonious SSVI model capturing this path-dependent behavior.
Fanfare 2Derivatives & VolatilityHervé Andrès et al.PDF
Decomposing forecasting losses into correlation versus scale components, the paper shows how to diagnose and stabilize dynamic-correlation volatility models using realized-volatility inputs.
Fanfare 2Derivatives & VolatilityHongfei Guo et al.PDF
Combining machine-learned forecasts of realized measures with dynamic conditional correlation models improves correlation matrix forecasts, producing valid predictions and beating realized-driver baselines across multiple horizons.
Fanfare 2Derivatives & VolatilityYongdeng XuPDF
The paper assesses how artificial intelligence affects monetary policy transmission and central bank reactions, finding AI could improve risk assessment and communication but may also amplify systemic vulnerabilities and herding dynamics.
Fanfare 3Macro-Finance & RatesLucia Esposito et al.PDF
Post-AI adoption, longer-maturity Indian bond yields show reduced sensitivity to expected inflation and money supply growth, while short-term yields exhibit heightened inflation sensitivity, suggesting structural transmission changes.
Fanfare 3Macro-Finance & RatesLekha Chakraborty and Prasanth C.PDF
U.S. green bond share declined from 1.7 to 0.6 percent after Trump's re-election and Paris Agreement withdrawal, with the greenium turning positive.
Fanfare 3Macro-Finance & RatesAlessandro Moro and Andrea ZaghiniPDF
Proposes a two-step estimator to recover portfolio-balance model parameters from Gaussian term structure models, identifying shocks to hedging risk premiums and risk-bearing capacity.
Fanfare 2Macro-Finance & RatesAntonio Diez de los RiosPDF
The paper examines how Federal Reserve balance sheet changes affect overnight Treasury repo markets and the transmission of monetary policy through money markets.
Fanfare 2Macro-Finance & RatesSriya Anbil et al.PDF
The research shows that both large-scale asset purchases and forward guidance appreciate foreign currencies against the dollar, with guidance having larger effects, especially during zero lower bound periods.
Fanfare 2Macro-Finance & RatesArisa ChantaraboonthaPDF
The study finds that geopolitical risk raises sovereign credit spreads in emerging markets, with threats having larger effects than acts, and responses shifting substantially after the Ukraine invasion.
Fanfare 2Macro-Finance & RatesFredy Gamboa and Jose Vicente RomeroPDF
Comparing FX forwards in identical currency pairs and maturities, the study finds large pricing variation across dealers reflecting clientele and pricing power rather than funding costs.
Fanfare 2Trading, Microstructure & ExecutionMarco Grotteria and Alex KontoghiorghesPDF
The study finds that UK borrowers shifted toward two-year fixed mortgages despite higher pricing after the 2022 rate shock, seeking flexibility and rate protection rather than minimizing immediate costs.
Fanfare 2Portfolio & AllocationPhilippe Bracke et al.PDF
