2026 Q1 Quarterly Research Archive¶
Records scoring at least 40 within the primary scope pass rule review and are published without additional manual review. This page does not validate author claims or provide investment advice.
- Coverage: 2026-01-01 to 2026-03-31
- Passed rule review: 15
- Sources: 8
Topic distribution¶
Domains¶
- volatility: 7
- hedging exposure risk: 5
- microstructure: 5
- execution costs: 4
- lifecycle infrastructure: 2
- option returns: 1
Methods¶
- financial ml: 7
- research methods: 6
Passed rule review¶
Reinforcement Learning for Jump‐Diffusions, With Financial Applications¶
- Published: 2026-03-17
- Source: Mathematical Finance
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper extends RL control algorithms from diffusion to jump-diffusion and claims invariant use for policy evaluation and option hedging, which is relevant to model-misspecification risk if its assumptions hold. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main author claims¶
- The authors claim RL for jump-diffusions in continuous time with entropy-regularized exploratory control, and provide careful treatment of jump components in exploratory dynamics. (
abstract:S1,abstract:S2,abstract:S3) - They claim one can reuse Jia and Zhou policy evaluation and Q-learning algorithms without pre-classifying diffusion versus jump-diffusion, with applications showing invariance in mean-variance portfolio selection and option hedging. (
abstract:S4,abstract:S5,abstract:S6)
Data, method, or discussion scope¶
Scope covers theoretical extension and claims of algorithmic invariance plus applications to mean-variance selection and option hedging; it omits discretization error, dataset specifics, jump-intensity estimation, and finite-sample robustness details. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
A key claim is unconditional transferability across model type, yet boundary conditions for misspecified dynamics or noise-heavy data are not defined. (abstract:S4, abstract:S5)
Finance-Informed Neural Network: Learning the Geometry of Option Pricing¶
- Published: 2026-03-11
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
FINN reframes pricing and hedging as learning a replication-consistent pricing operator, which is significant for reducing parametric misspecification risk and extending to illiquid markets, if trainability and calibration are sound. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S6, abstract:S7)
Main author claims¶
- The authors claim FINN is learned through a self-supervised replication objective based on dynamic hedging, integrating financial structure into the neural network. (
abstract:S1,abstract:S2) - They claim minimizing replication error recovers the arbitrage-free pricing operator, yields economically meaningful sensitivities, and remains stable in settings without reliable closed-form solutions. (
abstract:S3,abstract:S5,abstract:S4) - The authors claim FINN reconstructs implied-volatility surfaces closer to market values than Heston calibrations and can use historical spot prices to construct coherent option prices and Greeks for assets without listed options. (
abstract:S6,abstract:S7)
Data, method, or discussion scope¶
Scope includes theoretical objectives, replication consistency claims, recovery of Black-Scholes behavior, Heston robustness, implied-volatility surface performance versus Heston calibrations, and training on historical spot for illiquid assets. Hyperparameters and cost model integration are not provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S6, abstract:S7)
Main limitations¶
The claim of improved IV surface reconstruction lacks explicit metric definitions and windows, and abstract-level reporting does not specify transaction-cost treatment, liquidity thresholds, or walk-forward drawdown checks. (abstract:S4, abstract:S6, abstract:S9)
Uncertainty-Aware Deep Hedging¶
- Published: 2026-03-10
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
This work adds uncertainty quantification to deep hedging, which directly affects confidence-gating and risk-limit logic; CVaR improvements under uncertainty are operationally meaningful if replicable. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)
Main author claims¶
- The authors claim standard deep hedging lacks model confidence measures and introduce a deep ensemble of five LSTMs under Heston with proportional transaction costs to produce per-time-step uncertainty. (
abstract:S1,abstract:S2) - They also claim ensemble disagreement is predictive of performance and propose a CVaR-optimized blend with Black-Scholes delta, yielding statistically significant CVaR improvements over BS delta and Whalley-Wilmott in benchmark checks. (
abstract:S3,abstract:S4,abstract:S5)
Data, method, or discussion scope¶
Scope includes the gap statement, ensemble setup, comparison against BS and Whalley-Wilmott, paired bootstrap significance, and identified uncertainty driver (moneyness); training horizon and sensitivity analyses are not provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
Reported gains and significance lack explicit benchmark implementation details such as path construction and regime partitioning, leaving reproducibility under alternative transaction-cost models uncertain. (abstract:S3, abstract:S5)
The Role of Price‐Volatility Cojumps in Volatility Forecasting¶
- Published: 2026-03-09
- Source: Journal of Futures Markets
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper integrates price-volatility cojumps into HAR forecasting and separates upside/downside effects, which matters for event-triggered risk forecasting, provided high-frequency noise and sample-selection issues are controlled. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main author claims¶
- The authors claim to identify intraday price-volatility cojumps from high-frequency S&P 500 and VIX data and construct upside, downside, and asymmetric measures. (
abstract:S2) - They claim downside cojumps increase future volatility while upside cojumps reduce it, and adding these effects to HAR significantly improves out-of-sample forecasting. (
abstract:S3,abstract:S4) - The authors also report that recent price jumps become important volatility predictors when they coincide with volatility jumps. (
abstract:S5)
Data, method, or discussion scope¶
Scope covers high-frequency jump-volatility jump identification, directional decomposition, HAR integration, and out-of-sample improvement claims, plus the additional assertion that recent price jumps matter when co-jumps are present. (abstract:S2, abstract:S3, abstract:S4, abstract:S5)
Main limitations¶
Detection thresholds, noise filtering, and HAR configuration are not specified, so directional effects may be sensitive to high-frequency preprocessing choices. (abstract:S2, abstract:S3, abstract:S4)
Asymmetric Option Returns in China¶
- Published: 2026-03-09
- Source: Journal of Futures Markets
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper documents asymmetric option return patterns and behavior-based explanations in China’s options market, which can matter for period-specific bias monitoring, though behavioral attribution may be confounded with market microstructure factors. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main author claims¶
- The authors claim documented return asymmetries where call (put) returns are positive (negative) overnight but negative (positive) intraday in China’s options market. (
abstract:S1) - They provide an explanation based on demand pressure: retail gambling motives as primary, with insurance-driven demand for negative overnight put returns and attention-driven demand for intraday call asymmetry. (
abstract:S2,abstract:S3) - The authors report that the asymmetries persist despite institutional exploitation, leaving retail investors to bear losses as counterparties. (
abstract:S4)
Data, method, or discussion scope¶
Scope is limited to documented asymmetry patterns and demand-pressure explanations; no cross-sectional model specification, liquidity control, or interaction tests with frictions are included in the abstract. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main limitations¶
Behavioral explanations are inferential and lack explicit competing-mechanism tests such as liquidity frictions, funding constraints, and leverage effects. (abstract:S2, abstract:S3, abstract:S4)
Pricing and hedging for liquidity provision in Constant Function Market Making¶
- Published: 2026-03-02
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
This abstract proposes a coordinate transform for CFMMs that is directly relevant to consistent pricing and IL risk quantification in automated market-making systems, with practical risk limits likely hinging on calibration assumptions not shown in abstract-only form. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S6)
Main author claims¶
- The authors claim a switch to price and intrinsic liquidity coordinates with canonical parametrization that remains dimensionally consistent across CFMM trading functions. (
abstract:S1,abstract:S2) - They further claim this linear structure supports arbitrage-free pricing, delta hedging, and systematic risk management, with IL characterized as a weighted strip of vanilla options via Carr-Madan. (
abstract:S3,abstract:S4) - The authors report a volatility smile in Uniswap v3 ETH/USDC and Deribit data and use it to support their characterization of the risk-neutral fair value of liquidity provision. (
abstract:S6)
Data, method, or discussion scope¶
Scope includes methodological claims and empirical confirmation statements from Uniswap v3 ETH/USDC and Deribit, but no explicit error metrics, sample lengths, or outlier treatment are provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
Practical utility claims depend on selected market pairs and consistency assumptions; parameter estimation robustness and data-quality controls are not specified. (abstract:S6, abstract:S2, abstract:S5)
Learning to Optimally Stop Diffusion Processes, with Financial Applications¶
- Published: 2026-02-25
- Source: Management Science
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper recasts diffusion optimal stopping with unknown primitives as RL control, with implications for deploying learning-based free-boundary estimation in option and portfolio workflows. (abstract:S1, abstract:S5, abstract:S6, abstract:S11, abstract:S12, abstract:S13)
Main author claims¶
- The authors claim they transform the stopping problem into a stochastic control problem with two actions via penalized variational-inequality formulation, then randomize controls using Bernoulli distributions with entropy regularization for exploration. (
abstract:S5,abstract:S6) - They claim offline and online algorithms achieve high value-function accuracy and boundary characterization in finite-horizon American puts, Merton with transaction costs, and high-dimensional stopping extensions. (
abstract:S12,abstract:S13) - The authors state a policy-improvement theorem and claim fast convergence of the resulting policy iterations. (
abstract:S9,abstract:S10)
Data, method, or discussion scope¶
Scope includes the method pipeline, convergence statement, algorithmic design, and application areas stated in the abstract; network size, objective settings, and complexity requirements are not provided. (abstract:S1, abstract:S5, abstract:S6, abstract:S11, abstract:S12, abstract:S13)
Main limitations¶
Claims of fast convergence and high accuracy are not accompanied by convergence rates, training budgets, or noisy-data robustness conditions. (abstract:S11, abstract:S13)
Asian option valuation under price impact¶
- Published: 2026-02-21
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
The framework incorporates market impact and execution costs into Asian option valuation, linking pricing to execution policy in a way that matters for implementation where trading impact cannot be ignored. (abstract:S1, abstract:S2, abstract:S3, abstract:S7, abstract:S8, abstract:S10)
Main author claims¶
- The authors claim a tractable valuation framework for Asian options with market impact and execution costs, built from discrete quote-level dynamics to continuous-time coupled midpoint/impact systems. (
abstract:S1,abstract:S2,abstract:S3) - They claim two regimes are studied: an exogenous regime with a closed-form geometric Asian call expression under deterministic order-flow volatility and an endogenous regime cast as stochastic control, with a CRR-style tree Bellman solver for computation. (
abstract:S4,abstract:S5,abstract:S6,abstract:S7,abstract:S9) - The authors report impact-driven reservation bid-ask spreads under cost-based indifference; numerically, endogenous spreads grow super-linearly with impact, widen as execution costs fall, and narrow with faster mean reversion. (
abstract:S8,abstract:S10)
Data, method, or discussion scope¶
Scope includes regime decomposition, mathematical formulation and computational approach, bid-ask/indifference pricing interpretation, and qualitative experiment findings versus frictionless benchmarks; missing are calibration and numerical error bounds. (abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7, abstract:S8, abstract:S9, abstract:S10)
Main limitations¶
Experimental findings on modest exogenous effects and superlinear endogenous bid-ask behavior are not accompanied by parameter bounds, discretization error control, or computational cost scaling. (abstract:S10, abstract:S9)
Payment for Order Flow and Option Internalization¶
- Published: 2026-02-20
- Source: The Review of Financial Studies
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
This paper focuses on how DMM assignment and order-routing rules create internalization barriers in options wholesaling, which matters for market structure and execution-cost incentives rather than a pure price-model result. (abstract:S1, abstract:S3, abstract:S4)
Main author claims¶
- The authors claim option wholesalers specialize in purchasing and executing retail option order flow, with internalization occurring through auctions and the limit order book. (
abstract:S1,abstract:S2) - They further claim DMMs gain a five-contract advantage on LOB order internalization, and that DMM assignment and allocation rules generate an entry barrier in options wholesaling, supporting high option PFOF. (
abstract:S3,abstract:S4)
Data, method, or discussion scope¶
Scope covers mechanism-level claims about execution channels, DMM priority, and the structural inference about internalization barriers and profits; it does not provide cross-market samples or identification estimates in the abstract. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main limitations¶
Profit-protection and PFOF assertions lack magnitudes and counterfactual construction, so policy or welfare implications require additional identification detail. (abstract:S4)
Algorithmic Monitoring: Measuring Market Stress with Machine Learning¶
- Published: 2026-02-05
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
The abstract frames MSPI as a one-month-ahead equity-market stress probability, making it a candidate input for forward stress monitoring; threshold calibration and false-alarm governance are not provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S5)
Main author claims¶
- The authors claim to construct MSPI, a one-month-ahead probability index for U.S. market stress using cross-sectional stock information. (
abstract:S1,abstract:S2) - They claim MSPI tracks major stress episodes and improves discrimination and accuracy versus a parsimonious benchmark using lagged market return and realized volatility in out-of-sample tests. (
abstract:S3)
Data, method, or discussion scope¶
Reviewable scope includes the cross-sectional CRSP feature design, L1-regularized logistic regression with expanding real-time window, monthly horizon, and out-of-sample discrimination claims; full parameter and metric tables are not provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S5)
Main limitations¶
The key stress episodes and calibrated probabilities are not accompanied by thresholds, coverage, or false-positive and false-negative costs, so action rules cannot be inferred. (abstract:S3, abstract:S5)
Forecasting Crude Oil Volatility With Geopolitical Risk: The RSV–MIDAS–GPR Model and Its Economic Value¶
- Published: 2026-02-05
- Source: Journal of Futures Markets
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper embeds geopolitical risk into a realized-volatility MIDAS framework for crude oil futures forecasting, which is relevant to stress-aware volatility signals used for risk budgeting and risk controls, especially around turmoil states. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)
Main author claims¶
- The authors claim to propose an RSV–MIDAS–GPR integrated model that jointly models returns, realized volatility, and geopolitical risk for crude oil futures volatility. (
abstract:S1,abstract:S2) - They claim the model delivers a substantial out-of-sample forecast improvement, better high-volatility-state identification, and positive results in economic-value tests. (
abstract:S4,abstract:S5,abstract:S6) - The authors report a significant positive association between geopolitical risk and crude-oil futures volatility. (
abstract:S3)
Data, method, or discussion scope¶
Scope includes model definition, empirical statements on positive GPR-volatility relation, out-of-sample forecast improvement, high-volatility-state identification, and economic-value framing; it does not provide full specification details, sample splits, or metric formulas. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
Statistical significance, practical value, and outperformance claims are abstract-level; missing details on GPR construction, sample windows, metrics, and transaction costs prevent direct adoption. (abstract:S3, abstract:S4, abstract:S5, abstract:S6)
VIX Term Structure in the Rough Heston Model via Markovian Approximation¶
- Published: 2026-01-31
- Source: Journal of Futures Markets
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper replaces direct rough Heston simulation with a Markovian approximation and analytic gradients for VIX term-structure calibration; efficiency must be assessed together with the approximation-error tolerance. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main author claims¶
- The authors claim rough Heston direct simulation is inefficient, so they use a Markovian approximation with analytical VIX expressions that avoid simulation. (
abstract:S1,abstract:S2,abstract:S3,abstract:S4) - They claim analytical gradients further speed calibration and that in extensive daily VIX term-structure data the approach outperforms competing Heston-type-with-jumps models, yielding more reliable spot-volatility estimates. (
abstract:S5,abstract:S6)
Data, method, or discussion scope¶
Scope is limited to abstract-level claims of efficiency and comparative fit/estimate reliability; it lacks approximation error bounds, truncation levels, and failure-case definitions. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
Competing sets and significance thresholds for outperformance are not specified; reliability is not translated into explicit error diagnostics, and stress behavior under liquidity shocks is not shown. (abstract:S6, abstract:S4, abstract:S2, abstract:S3)
Machine learning for option pricing: an empirical investigation of network architectures¶
- Published: 2026-01-29
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
This preprint studies how architecture choice affects accuracy and training time in option-pricing tasks, informing model-family selection while leaving parameter-budget fairness and training stability unspecified. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)
Main author claims¶
- The authors claim that much of the literature uses plain feedforward networks, and they investigate whether architecture choice affects accuracy and training time. (
abstract:S1,abstract:S2,abstract:S3) - They claim generalized highway networks perform best under MSE and training-time criteria for Black-Scholes/Heston pricing within their budgets, while simplified DGM gives lowest error for transformed implied-volatility tasks. (
abstract:S4,abstract:S5)
Data, method, or discussion scope¶
The scope includes architecture comparison claims, metrics, parameter budgets, and a mention of real-market-data experiments for implied-volatility, without explicit hyperparameters or significance metrics. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)
Main limitations¶
Best-performance claims depend on unspecified budget and task definitions, and no seeding/stability or cross-asset transfer procedures are reported. (abstract:S4, abstract:S6, abstract:S7, abstract:S3)
Cheap Options Are Expensive¶
- Published: 2026-01-16
- Source: The Review of Asset Pricing Studies
- Publication status:
peer_reviewed - Original source: Open original source
Why it matters¶
The paper reports that low-price-stock options are relatively expensive versus high-price stocks and provides split evidence via stock splits and mini contracts; this matters for market microstructure and demand-pressure interpretations. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main author claims¶
- The authors report weekly relative shortfalls for low-price-stock options of 63 basis points for calls and 36 for puts, using delta-hedged returns. (
abstract:S1) - They claim natural experiments on stock splits, mini indices, and mini contracts support this effect, attributing it to retail skewness preference and divergence of opinion, with limits-to-arbitrage and market-making effects insufficient to fully explain it. (
abstract:S2,abstract:S3,abstract:S4)
Data, method, or discussion scope¶
Reviewable scope covers the reported effect and natural experiments plus mechanism wording; no trading costs, sample composition, or out-of-sample error structure is provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)
Main limitations¶
Natural-experiment support is claimed without matching criteria, and attributed mechanisms are not decomposed from liquidity, hedging costs, or maker inventory constraints. (abstract:S2, abstract:S4, abstract:S3)
American options valuation in time-dependent jump-diffusion models via integral equations and characteristic functions¶
- Published: 2026-01-08
- Source: arXiv Quantitative Finance History
- Publication status:
preprint - Original source: Open original source
Why it matters¶
The preprint proposes a semi-analytical integral-equation route for American options in time-inhomogeneous jump models, relevant to numerical treatment of complex exercise boundaries and computational efficiency. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main author claims¶
- The authors claim the approach applies to diffusion, jump-diffusion, and Levy models and solves exercise boundaries via Volterra integral equations of the second kind. (
abstract:S2,abstract:S3) - They claim the framework addresses cases without closed-form transition densities via characteristic functions/COS and generalizes to multidimensional diffusions; numerical examples are presented as efficient and robust with industrial relevance. (
abstract:S4,abstract:S5,abstract:S6)
Data, method, or discussion scope¶
Scope is abstract-level methodological claims and numerical example mention; no explicit error bounds, complexity orders, or multidimensional feasibility limits are provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)
Main limitations¶
Claims of superiority and industrial applicability are not paired with concrete baselines, accuracy thresholds, or failure behavior, so they are insufficient for choosing an implementation. (abstract:S3, abstract:S6, abstract:S5)