2026 Q2 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: 2026-04-01 to 2026-06-30
• Passed rule review: 28
• Sources: 9

Topic distribution

Domains

• volatility: 21
• microstructure: 6
• execution costs: 6
• option returns: 2
• hedging exposure risk: 2
• portfolio construction risk transfer: 1
• lifecycle infrastructure: 1

Methods

• research methods: 13
• financial ml: 10

Facets

• instrument index options: 4
• instrument etf options: 4
• instrument vix options: 2
• instrument single stock options: 1
• horizon 0dte: 1
• horizon weekly: 1
• horizon short dated: 1
• exposure dispersion: 1
• structure iron condor: 1

Passed rule review

Regime-Conditional Distributional Comparison of Trading Strategies: A GAMLSS/ZAGA Framework Applied to the S&P 500

• Published: 2026-06-30
• Source: arXiv Quantitative Finance
• Publication status: preprint
• Original source: Open original source

Why it matters

The paper proposes a distributional comparison framework instead of single-number strategy ranking, which matters when strategy performance is regime-dependent. (abstract:S1, abstract:S2, abstract:S6)

Main author claims

• The authors argue conventional single-metric comparisons can mask market-state dependence. (abstract:S1, abstract:S2)
• They run 146 out-of-sample folds for SVMP versus BH IR* and model them with a GAMLSS/ZAGA framework. (abstract:S3, abstract:S4)
• They claim dominance is regime-specific, tested via ΔE/ΔVar and parametric bootstrap over six representative market regimes. (abstract:S5, abstract:S6, abstract:S7)

Data, method, or discussion scope

Scope includes fold count, strategy definitions, GAMLSS/ZAGA modeling framework, and regime-based testing design; it does not include full parameterization and full per-regime numeric tables. (abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)

Main limitations

IR* definition details, bootstrap setup, thresholds, and failure scenarios are not fully specified, limiting interpretability and reproducibility checks. (abstract:S4, abstract:S5, abstract:S7)

The cross-section of individual equity option returns

• Published: 2026-06-25
• Source: Journal of Empirical Finance
• Publication status: peer_reviewed
• Original source: Open original source

Why it matters

The study offers an interpretable sparse-model baseline for individual-equity option returns, examining both forecast performance and whether selected characteristics survive resampling and reselection. It is useful for feature evaluation in delta-hedged option-return research, not as a model of two-leg terminal payoffs, max-loss-normalized returns, or an executable trading strategy. (full_text:S27, full_text:S136, full_text:S138, full_text:S179, full_text:S385)

Main author claims

• The authors start with 100 stock/firm and 30 option characteristics, remove 22 highly collinear variables, and apply LASSO to monthly cross-sectional ranks. They repeat selection and post-selection Fama–MacBeth estimation across 1,000 resamples of whole months. Historical-minus-implied volatility (hv_iv) and idiosyncratic volatility are selected in every resample, and turnover in roughly 98%–100%. Conditional on the other selected predictors, hv_iv and turnover have positive associations with subsequent delta-hedged returns, while idiosyncratic volatility has a negative association. (full_text:S27, full_text:S108, full_text:S137, full_text:S138, full_text:S230, full_text:S232, full_text:S256, full_text:S271, full_text:S273, full_text:S282)
• The authors report: In Table 5, the authors reselect and re-estimate models in rolling 120-month training windows and report 159 monthly out-of-sample forecasts. Adaptive LASSO with BIC gives realized-on-predicted slopes of 0.979 for calls and 1.045 for puts, with cross-sectional regression R-squared near 2.1% for both. Predicted-return decile high-minus-low portfolios have annualized Sharpe ratios of 3.351 and 4.228. The higher reported maxima of 3.929 and 4.723 use other penalty settings; they should not be conflated with BIC performance or interpreted as evidence of returns net of full execution costs. (full_text:S398, full_text:S399, full_text:S402, full_text:S408, full_text:S409, full_text:S410, full_text:S416, full_text:S177)
• The authors' double sorts associate larger hv_iv return spreads with information frictions, demand pressure, and illiquidity, while idiosyncratic-volatility spreads strengthen under several arbitrage and hedging-friction proxies. Turnover behaves differently, with stronger patterns in volatile and smaller-stock segments rather than a consistent arbitrage-cost pattern across liquidity proxies. These are conditional associations compatible with proposed mechanisms, not causal identification of those mechanisms. (full_text:S460, full_text:S462, full_text:S480, full_text:S486, full_text:S528, full_text:S529, full_text:S530)

Data, method, or discussion scope

The main paper uses US individual-equity options from 1996–2019, combining OptionMetrics, CRSP, Compustat, and I/B/E/S, with approximately 200,000 call and 160,000 put contract-month observations. It retains near-ATM calls and puts per stock where available. Actual initial maturities are 37–52 days; positions are held for approximately one month with daily delta rebalancing, not to expiry. Returns divide hedged gains by the absolute initial net cash position |O₀−Δ₀S₀|, not the premium, margin, or maximum loss. Annual and quarterly accounting inputs are lagged six and four months. First online publication was 2026-06-25, not the end of the empirical sample. (full_text:S21, full_text:S141, full_text:S144, full_text:S145, full_text:S159, full_text:S170, full_text:S174, full_text:S175, full_text:S177, full_text:S179, full_text:S205, full_text:S218)

Main limitations

Table 6 fixes characteristics using the full-sample bootstrap and only rolls coefficient estimation, giving feature choice a hindsight advantage; it is not the rolling-reselection design of Table 5. The main text does not establish whether Table 5 also repeats VIF pruning within each training window or how CV folds respect time. Reported R-squared comes from cross-sectional regressions with an intercept, not forecast-error OOS R-squared against zero returns; a slope near one alone does not establish unbiasedness. Returns use bid/ask midpoints and deduct risk-free financing, but the main paper does not report portfolios net of spreads, daily hedge execution, stock borrowing, margin constraints, and impact. Resampling whole months preserves within-month dependence, not serial dependence across months, and does not automatically establish coverage for nonregular post-selection intervals. Double sorts do not identify causality, and the historical ATM monthly sample does not establish validity for today's full option surface, 0DTE, or two-leg combinations. All 25 supplied pages were read; the separate Online Appendix was unavailable, and the authors' data and code were not independently replicated. (full_text:S52, full_text:S53, full_text:S137, full_text:S140, full_text:S145, full_text:S175, full_text:S177, full_text:S179, full_text:S205, full_text:S230, full_text:S231, full_text:S239, full_text:S386, full_text:S404, full_text:S405, full_text:S411, full_text:S441, full_text:S460, full_text:S553)

How to Intraday Backtest Double Calendars | Driven By Data Ep. 137

• Published: 2026-06-23
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

The video description maps an intraday double-calendar backtest workflow—data frequency, entry, costs, expiries, and result checks—making it useful as a tool-discovery entry point. (description:S1, description:S3, description:S4, description:S5)

Main author claims

• The episode description says the ORATS backtester uses one-minute data, smoothing, and Greeks, and discusses the 9:34 default entry and differences between weekly and daily expiries. (description:S4)
• The episode also covers slippage, commissions, margin versus notional returns, trade logs, and full-history checks. (description:S3, description:S5)

Data, method, or discussion scope

The basis is an episode description and timestamps, not a transcript or reproducible backtest; it describes features and a teaching agenda without extraction rules, parameter tables, or performance results. (description:S1, description:S3, description:S4, description:S5, description:S6)

Main limitations

The content comes from the tool provider and explicitly disclaims accuracy and outcomes and notes that past performance is not predictive; the description cannot validate strategy returns or live-execution fitness. (description:S10, description:S13, description:S14, description:S25, description:S28)

How Chicago Became the World’s Options, Vol, and Derivatives Capital (with Cboe’s Rob Hocking & Mandy Xu)

• Published: 2026-06-18
• Source: The Derivative by RCM Alternatives
• Publication status: unknown
• Original source: Open original source

Why it matters

The episode provides institutional-history and market-structure context (Chicago, zero-DTE, volatility, AI products) useful for framing, but it is primarily narrative rather than directly quantitative evidence. (description:S2, description:S5, description:S7)

Main author claims

• The authors report: The episode describes Chicago’s evolution from agricultural hedging roots to the SPX and VIX risk-transfer center. (description:S2, description:S3)
• The authors report: It discusses zero-DTE liquidity dynamics, VIX interpretation, AI-driven dispersion, and emerging productization in derivatives. (description:S6, description:S7)
• The authors report: The episode states that the content is for informational purposes and not legal, business, or tax advice. (description:S13, description:S14, description:S15)

Data, method, or discussion scope

Evidence scope is narrative history and participants’ views; it does not include auditable trading statistics or causal policy tests. (description:S2, description:S3, description:S4, description:S6)

Main limitations

Many conclusions are experiential and narrative, lacking time-series comparisons and quantifiable indicators, with conclusions dependent on participant perspectives. (description:S3, description:S10, description:S11)

Reciprocal Return Risk Premium and Option Returns

• Published: 2026-06-18
• Source: Journal of Futures Markets
• Publication status: peer_reviewed
• Original source: Open original source

Why it matters

The paper links reciprocal return risk premium with option cross-sectional returns, making sample definition, decomposition stability, and risk-premium interpretation critical for use. (abstract:S1, abstract:S2, abstract:S3)

Main author claims

• The authors claim reciprocal return risk premium significantly predicts cross-sectional option returns. (abstract:S1)
• They emphasize the central role of ex post volatility risk premium in the claimed relation. (abstract:S2)
• The authors report: The positive relation is attributed mainly to a natural return-variation component under the risk-neutral decomposition. (abstract:S3)

Data, method, or discussion scope

Evidence includes the relation statement, theoretical interpretation, and attribution claim, but not explicit significance thresholds, asset-screening rules, or multi-market robustness evidence. (abstract:S1, abstract:S2, abstract:S3)

Main limitations

Absence of sample and regression details makes it difficult to map claimed significance and centrality into decision-grade constraints. (abstract:S1, abstract:S2)

From Arbitrage Removal to Density Extraction: A Model-Free Framework for Short-Dated Options

• Published: 2026-06-11
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The framework attempts to use market constraints instead of full model assumptions for short-dated chains; the key issue is whether this improves robustness and operational usability. (abstract:S2, abstract:S4, abstract:S7, abstract:S9)

Main author claims

• The authors state that nearing expiry and wide bid-ask spreads undermine mid-quote-based standard methods, motivating a constraint-first alternative. (abstract:S2, abstract:S3)
• They propose ARIES for executable static-arbitrage removal and SEDEx for bid-ask constrained smooth entropic density recovery. (abstract:S6, abstract:S7)
• The authors report: The pipeline is reported to run quickly on synthetic Heston and short-dated SPX data and is used to build short-dated IV smiles. (abstract:S8, abstract:S9, abstract:S10)

Data, method, or discussion scope

Scope covers the stated motivation from stale/aspread issues, the two-stage ARIES/SEDEx pipeline, and qualitative speed/robustness claims plus an IV-smile application. (abstract:S2, abstract:S3, abstract:S4, abstract:S6, abstract:S7, abstract:S8, abstract:S10)

Main limitations

Parameter grids, stopping rules, failure cases, and error propagation are not specified, so reported speed and robustness cannot be reconciled to engineering acceptance criteria. (abstract:S7, abstract:S9, abstract:S10)

Signs of a Market Top from the Options Market | Driven By Data Ep. 135

• Published: 2026-06-09
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

This item is about warning logic, where the key question is whether indicator narratives can be converted into executable caution signals rather than remaining SPY-specific interpretation. (description:S2, description:S4, description:S7)

Main author claims

• The authors report: The episode claims option-derived signals can detect fragility not always reflected in stock charts. (description:S2)
• The authors report: It presents rising IV, higher SPY relative volatility, and falling contango as stress-warning signals. (description:S4)
• The authors report: The episode states its aim is earlier caution-zone detection, not precise top timing. (description:S6, description:S7)

Data, method, or discussion scope

Evidence scope is a SPY-oriented demonstration and warning-signal explanation; it excludes threshold rules, false-positive metrics, cross-market validation, and executable backtests. (description:S2, description:S3, description:S4, description:S7)

Main limitations

No quantified alert rules or error control are provided, and the transition to concrete position adjustments is unspecified. (description:S4, description:S7)

Volatility Forecasting and Return Prediction under Market Regimes: Evidence from High-Frequency Chinese Equity Data

• Published: 2026-06-08
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This preprint joins regime identification, volatility forecasting, return prediction, and implementation in one pipeline, while explicitly showing that better forecasts do not automatically become post-cost strategy value. (abstract:S5, abstract:S6, abstract:S7)

Main author claims

• On high-frequency CSI 300 data from 2005–2023, the authors report that regime-aware volatility models outperform baseline HARQ across forecast metrics. (abstract:S2, abstract:S5)
• The authors find weak return predictability concentrated in low-volatility regimes and state that volatility scaling, gating, thresholds, and turnover controls can improve defensive economic performance. (abstract:S6, abstract:S7, abstract:S8)

Data, method, or discussion scope

The abstract describes a two-stage HARQ / Markov-switching GJR-GARCH and XGBoost framework with walk-forward out-of-sample estimation, but gives no result table, cost function, or gating parameters. (abstract:S2, abstract:S3, abstract:S4)

Main limitations

Evidence comes from one Chinese equity index and the return signal is weak and regime-dependent; economic improvement may be sensitive to regime definitions, thresholds, turnover, and cost assumptions. (abstract:S2, abstract:S6, abstract:S7)

The Dynamic Extreme Comovement Between Options Market Ambiguity and Implied Volatility

• Published: 2026-06-07
• Source: Journal of Futures Markets
• Publication status: peer_reviewed
• Original source: Open original source

Why it matters

The study proposes an extreme-comovement signal between OMA and implied volatility; the key issue is whether this leading-indicator claim is robust and tradable. (abstract:S1, abstract:S2, abstract:S3)

Main author claims

• The authors claim to model extreme comovement between OMA and implied volatility using a symmetrized Joe-Clayton copula with recursive estimation. (abstract:S1)
• They claim AEC supplies incremental information about future uncertainty beyond OMA and implied volatility. (abstract:S2)
• They further claim AEC predicts downturns in and out of sample and improves strategy performance versus benchmarks. (abstract:S3, abstract:S4)

Data, method, or discussion scope

Evidence covers the model family and high-level predictive/performance claims, but not full significance detail, cost assumptions, or complete sample definitions. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main limitations

The leading-indicator and outperformance language is largely declarative and lacks reproducible error accounting plus execution-friction treatment. (abstract:S3, abstract:S4)

The Breakthrough Intraday Backtester for All Symbols | Driven By Data Ep. 134

• Published: 2026-06-02
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

This item hinges on whether the claimed functionality of intraday backtesting materially changes what can be tested in options research, rather than the presentation quality itself. (description:S1, description:S2, description:S6)

Main author claims

• The authors report: The episode claims intraday backtesting is available for any symbol with weekly options, presented as more than just adding data points. (description:S2, description:S7)
• The authors report: It cites one-minute data, ORATS SMV processing, and comparison of intraday versus end-of-day results. (description:S4, description:S6)
• They claim this enables a broader class of options strategies than traditional EOD backtesting. (description:S7, description:S8)

Data, method, or discussion scope

Scope includes capability claims, data frequency claims, and intraday-vs-EOD comparison logic, with no detailed benchmark table or execution parameters. (description:S2, description:S3, description:S4, description:S6, description:S9)

Main limitations

The coverage and reliability of the any-symbol claim are not demonstrated, and no error decomposition across samples/liquidity regimes is shown. (description:S1, description:S2, description:S6)

Tom Sosnoff on Options Trading, Entrepreneurship, AI, & Risk Management | AlphaMinds Fintech Leaders

• Published: 2026-05-30
• Source: Quantopian Webinars
• Publication status: unknown
• Original source: Open original source

Why it matters

The interview provides practitioner perspective from a market participant; value lies in framing assumptions and decision heuristics, with limited generalizability due to its experiential nature. (description:S2, description:S4, description:S7)

Main author claims

• The authors report: The episode covers Tom Sosnoff’s practitioner perspective across his CBOE, Thinkorswim, and Tastytrade trajectory. (description:S2, description:S3)
• The authors report: It covers execution, position sizing, and AI-informed education themes, but largely as opinion-led discussion. (description:S4)
• The authors report: The episode explicitly states the material is educational and not investment advice or a recommendation for specific products. (description:S7, description:S8, description:S9)

Data, method, or discussion scope

Evidence scope is interview topic coverage and disclaimer text only; no reproducible models, backtests, or measured risk-effect identification are provided. (description:S2, description:S3, description:S4, description:S7)

Main limitations

No explicit sample boundary or empirical method is provided, so applicability and transferability of opinions are limited. (description:S4, description:S10, description:S11)

Inspectable Neural Markov Models for Non-Stationary Time Series

• Published: 2026-05-29
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The method targets sparse-data settings while preserving interpretable transition structure, which is relevant for state modeling and regime analysis in stressed markets. (abstract:S2, abstract:S3, abstract:S5)

Main author claims

• The authors propose neural-network parameterization of stochastic matrix manifolds to estimate time-inhomogeneous Markov chains under sparse data. (abstract:S2, abstract:S3)
• They claim realized-volatility conditioning gives a more internally consistent Markov structure and improves held-out likelihood in 9 of 10 assets. (abstract:S4)
• They also claim transition-probability homogenization in high-volatility regimes and argue the explicit matrices improve geometric interpretability. (abstract:S5)

Data, method, or discussion scope

Scope includes rationale for non-stationary modeling, the neural manifold parameterization idea, and abstract quantitative headlines around held-out likelihood and a 5.6% CK discrepancy reduction. (abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main limitations

Training constraints, sample partitioning, and significance reporting are incomplete, making reproducibility of the 5.6% improvement and geometric claims difficult to verify. (abstract:S3, abstract:S4, abstract:S5)

Using AI + ORATS CLI to Build an Options Research Agent | Driven By Data Ep. 133

• Published: 2026-05-27
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

This companion episode reiterates the ORATS CLI narrative; its incremental value depends on reproducible and auditable workflow boundaries that the description does not establish. (description:S1, description:S3, description:S6)

Main author claims

• The authors report: The episode likewise claims ORATS CLI and AI can support a research-agent pattern rather than single prompts. (description:S1, description:S2)
• The authors report: The workflow is framed as dependent on permissions, installation, token access, and local-versus-remote execution boundaries. (description:S2, description:S5, description:S6)
• The authors report: The episode suggests persistence should be supported through structured summaries and personalized preference memory. (description:S3, description:S5, description:S6)

Data, method, or discussion scope

Scope is demonstration and workflow description: it covers research-agent framing, permission topics, and scheduling considerations, but not reproducible logs or benchmark tables. (description:S2, description:S3, description:S5, description:S6)

Main limitations

The item lacks implementation metrics (failure rate, latency, credential-expiry policy), and much of the content overlaps prior material, limiting incremental verifiability. (description:S5, description:S6, description:S7)

Using AI and the ORATS CLI to Build an Options Research Agent | Driven By Data Ep. 133

• Published: 2026-05-26
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

The episode shows an ORATS-AI workflow for an options research agent, relevant to moving from ad-hoc queries toward repeatable research operations; auditability and cross-environment reproducibility remain unresolved. (description:S1, description:S2, description:S4)

Main author claims

• The authors report: The episode claims the ORATS CLI can be paired with AI tools to build an options research agent beyond one-off question answering. (description:S1, description:S2)
• The authors report: It emphasizes workflow differences across live, delayed, and end-of-day contexts and positions the CLI relative to the standard dashboard. (description:S2)
• The authors report: The episode includes parallel sub-agents, session memory, structured summaries, and scheduling elements in the workflow. (description:S3, description:S4, description:S5, description:S6)

Data, method, or discussion scope

Evidence is confined to functional narration and workflow demonstration; it does not include reproducible configuration manifests, benchmark metrics, or quantified controls for security/compliance. (description:S1, description:S2, description:S3, description:S5, description:S6)

Main limitations

No reproducible permission-change logs, error-rate metrics, throughput boundaries, or cross-environment repeatability evidence are provided; the material is directional and lacks acceptance criteria. (description:S5, description:S6, description:S30)

Memory, Roughness, and Information Persistence in Financial Markets: A Structural Approach to Volatility Forecasting

• Published: 2026-05-22
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The work combines long-memory, roughness, and persistence features in a volatility-prediction framework, asking whether they add out-of-sample information beyond conventional volatility predictors during stress. (abstract:S1, abstract:S2, abstract:S8)

Main author claims

• The authors claim to combine long-memory dynamics, rough-volatility behavior, and persistence features in equity volatility modeling. (abstract:S1, abstract:S2)
• They report GP-H and local-Whittle memory estimates on 115 S&P 500 constituents from November 2001 to April 2026, with significance broadly documented. (abstract:S3, abstract:S4)
• They further claim moderate out-of-sample improvements from persistence aggregates, with stronger gains at longer horizons and during stress regimes, while explicitly declining to claim structural identification of the underlying economic mechanisms. (abstract:S6, abstract:S7, abstract:S8, abstract:S9)

Data, method, or discussion scope

Reviewable scope includes sample definition, memory parameter estimates, comparison with HAR/HAR-X in out-of-sample design, and reported stress/high-volatility gains; it does not include full test statistics, execution assumptions, or downstream economic interpretation. (abstract:S3, abstract:S4, abstract:S6, abstract:S7, abstract:S8)

Main limitations

The abstract omits regression specifications, hyperparameter grids, robustness-window design, and a profit-and-loss realization layer, so the statistical improvement cannot be converted directly into an executable rule. (abstract:S2, abstract:S6, abstract:S8)

SANOS Smooth strictly Arbitrage-free Non-parametric Option Surfaces

• Published: 2026-05-22
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This nonparametric construction claims smooth, strictly arbitrage-free surfaces with LP calibration and simple constraints, which is relevant to consistent surface construction if computational and scaling behavior is validated. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main author claims

• The authors claim a numerically efficient nonparametric method to represent option price surfaces that are smooth and strictly arbitrage-free across time and strike, with a smooth generalization of linear interpolation. (abstract:S1, abstract:S2)
• They claim calibration is formulated as a linear program with bid-ask penalties or inequalities, yielding strictly positive discrete local-volatility variables and arguing this is a first construction with only trivial positivity constraints. (abstract:S3, abstract:S4, abstract:S5)

Data, method, or discussion scope

Scope includes the method properties, LP calibration formulation with bid-ask penalties/inequalities, positivity parameterization, and illustration on S&P 500 options. Scalability benchmarks and large-scale stress tests are not included. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main limitations

No complexity growth curves, grid-selection details, or off-S&P500 error comparisons are provided, making scalability and numerical stability under stress difficult to assess. (abstract:S3, abstract:S5, abstract:S6)

What Does Deep Hedging Actually Learn? Delta Corrections, Regime Fragility, and Symbolic Distillation

• Published: 2026-05-20
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The paper asks what learned hedges do and when they fail, emphasizing auditability and regime fragility; symbolic distillation may turn black-box policies into auditable rules, but does not establish production readiness. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main author claims

• The authors claim empirical deep hedging for S&P 500 options is studied against a daily-updated Black-Scholes delta benchmark. (abstract:S1, abstract:S3)
• They claim walk-forward TD3 agents learn systematic delta haircuts, improve reward/downside-variance in many periods, but are regime-fragile; symbolic distillation preserves much of the advantage while inheriting fragility. (abstract:S4, abstract:S5, abstract:S6)

Data, method, or discussion scope

Scope includes the empirical setup, explainability focus, walk-forward period, observed failure modes, and symbolic distillation claims; cut rules and exact threshold definitions are not specified. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main limitations

Claims of partial improvement and regime failures are directional without explicit risk-threshold definitions or broader market replication. (abstract:S4, abstract:S5, abstract:S6)

Volatility Surface Reconstruction using Deep Learning under No-Arbitrage Constraints

• Published: 2026-05-20
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This study reconstructs implied-volatility surfaces from sparse noisy quotes with no-arbitrage constraints, which is important for quote-clean environments and model selection under incomplete markets. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main author claims

• The authors claim to reconstruct implied-volatility surfaces under no-arbitrage constraints from sparse and noisy quotes and benchmark multiple network architectures against classical SVI parameterizations. (abstract:S1, abstract:S2)
• They further claim Transformer and U-Net perform strongly, especially in sparse regimes, and that soft arbitrage penalties reduce arbitrage violations with moderate effect on reconstruction error. (abstract:S3, abstract:S4)

Data, method, or discussion scope

The scope includes sparse-noisy quote reconstruction, architecture set, sparse-regime performance claims, and explicit accuracy versus arbitrage-consistency trade-off analysis. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main limitations

Exact regularization grids, compute costs, and out-of-domain error bounds are absent, as are operational metrics for risk and latency. (abstract:S3, abstract:S4)

Customizing Your Options Backtests With AI | Driven By Data Ep. 132

• Published: 2026-05-19
• Source: ORATS Video
• Publication status: unknown
• Original source: Open original source

Why it matters

This item is a tool-demo episode description about AI-assisted options backtesting workflow; its value is operational process guidance, not empirical pricing or hedging evidence. (description:S1, description:S2, description:S3, description:S4, description:S5, description:S6, description:S7, description:S8)

Main author claims

• The authors report: The description states the ORATS AI backtest tool can convert plain-language ideas into backtests and that review/refinement materially affects outcomes. (description:S1, description:S2, description:S4)
• The authors report: It also describes a workflow from input editing and exit logic through optimization, paper trading, and autotrade, while repeatedly stating informational-only and significant-risk caveats. (description:S7, description:S6, description:S8, description:S9, description:S10, description:S11)

Data, method, or discussion scope

The scope is a workflow description and feature listing, with no formal performance metrics; the scope is mainly bounded by educational framing and risk disclaimers. (description:S1, description:S2, description:S3, description:S4, description:S5, description:S6, description:S7, description:S8, description:S9, description:S10, description:S11)

Main limitations

The video description does not provide reproducible performance evidence, cost modeling, or out-of-sample evaluation; claims are time-sensitive and demo-centric. (description:S8, description:S9, description:S12, description:S13)

Synthetic American Option Pricing via Jump-HMM-Driven Heston Implied Volatility

• Published: 2026-05-13
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This work addresses the circular dependency between implied-volatility and option prices by generating implied volatility from a structural return model, which is relevant for synthetic data quality and coherent risk-testing pipelines. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7, abstract:S8)

Main author claims

• The authors claim synthetic data generation breaks circularity by deriving implied-vol paths from Jump-HMM return paths through a modified Heston process and pricing American options with a recombining binomial lattice. (abstract:S1, abstract:S2, abstract:S3)
• They claim smile, skew, and term structure emerge without external calibration, with a hierarchical parametric+shared+sector neural surrogate for multi-sector ladders and event-driven generalization in temporal holdout. (abstract:S4, abstract:S5, abstract:S6, abstract:S7)
• The authors report joint simulation of path-conditional implied volatility, finite-difference American Greeks, and terminal short-premium PnL on real near-the-money options, followed by a second-underlying robustness run; the implementation is released as an open-source Julia package. (abstract:S7, abstract:S8)

Data, method, or discussion scope

Scope includes the broken-circularity claim, Jump-HMM and modified Heston pipeline, hierarchical calibration (parametric + neural surrogates), temporal holdout findings, and the open-source Julia package statement. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7, abstract:S8)

Main limitations

The pipeline depends on state, mood and surrogate design assumptions with limited reproducible specification, making risk diagnostics and failure-mode tracing harder. (abstract:S3, abstract:S5, abstract:S7)

Robust financial calibration: a Bayesian approach for neural SDEs

• Published: 2026-05-09
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This introduces Bayesian calibration for neural SDEs with posterior inference, adding explicit uncertainty quantification versus point estimates; this is important for model-risk-aware calibration workflows. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main author claims

• The authors claim a Bayesian neural SDE calibration framework with prior and likelihood specification and a global universal approximation result via Barron-type estimates. (abstract:S1, abstract:S2)
• They claim the posterior acts as a mixture over classical neural SDE models yielding robust implied-volatility-surface bounds, with Langevin-type sampling used for numerical optimization. (abstract:S3, abstract:S5)
• The authors use both historical time-series and option-price data and state that this requires learning the change between risk-neutral and historical measures. (abstract:S4)

Data, method, or discussion scope

Scope includes joint use of historical and option data, prior/likelihood specification, posterior mixture interpretation, measure-change requirement, and Langevin sampling for optimization. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main limitations

Robust bounds are claimed without prior-sensitivity and convergence diagnostics, and measure-change estimation error is not detailed. (abstract:S4, abstract:S5)

Proactive Market Making and Liquidity Analysis for Everlasting Options in DeFi Ecosystems

• Published: 2026-05-07
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The paper analyzes everlasting options in DeFi with a proactive market maker, including funding fees and transaction costs, which is relevant for liquidity design and incentive alignment in fragmented venues. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main author claims

• The authors claim everlasting options address rolling-contract and liquidity-fragmentation issues in DeFi, and they analyze their markets using a dynamic proactive market maker model. (abstract:S1, abstract:S2)
• They claim simulation and modeling indicate LPs can target positive net PnL through hedging even in low-liquidity/high-cost settings, and they describe LP incentives and trader benefits of everlasting options. (abstract:S3, abstract:S4, abstract:S5)

Data, method, or discussion scope

Scope covers product rationale, proactive market-making modeling, funding-fee/transaction-cost behavior, and simulation-based PnL statements; no simulation setup details, risk limits, or capital constraints are provided. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main limitations

The positive-net-profit claim is simulation-based and omits execution delays, model risk, and liquidation risk, so it cannot be extrapolated to a live protocol. (abstract:S4, abstract:S5)

A Geometry-Aware Residual Correction of Hagan's SABR Implied Volatility Formula

• Published: 2026-05-07
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The approach keeps a structured SABR backbone and learns residual corrections, potentially balancing interpretability and speed for calibration systems if residual behavior is validated across regimes. (abstract:S1, abstract:S2, abstract:S3, abstract:S6, abstract:S7, abstract:S8, abstract:S9)

Main author claims

• The authors claim a hybrid method that augments neural inputs with SABR-geometric features and trains the network to learn residual error versus Hagan’s approximation rather than raw implied volatility. (abstract:S1, abstract:S2, abstract:S5, abstract:S6)
• They further claim improved accuracy and robustness over analytical and standard neural approaches under realistic and stressed regimes, while remaining lightweight and structurally consistent for real-time pricing and calibration. (abstract:S7, abstract:S8, abstract:S9)

Data, method, or discussion scope

Scope includes the hybrid design, residual-learning target, experimental claims in regular and stressed settings, and practical calibration suitability; specific metrics, baselines, and runtime benchmarks are absent. (abstract:S1, abstract:S2, abstract:S3, abstract:S6, abstract:S7, abstract:S8, abstract:S9)

Main limitations

No explicit error metrics or stress-scenario definitions are provided for the reported gains, limiting tolerance analysis under calibration perturbations. (abstract:S8, abstract:S9)

American Options Pricing under Heston Model via Curriculum Learning in Coupled PINNs

• Published: 2026-05-01
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The method applies coupled PINNs to jointly learn American option prices and free boundaries under Heston, with curriculum learning and adaptive resampling for training stability, which is operationally relevant for PDE-boundary problems. (abstract:S1, abstract:S2, abstract:S3, abstract:S5, abstract:S6, abstract:S7)

Main author claims

• The authors claim Heston American options lack closed form, so they propose coupled PINNs to jointly learn option price and free boundary, with curriculum learning and adaptive resampling for stable training. (abstract:S3, abstract:S4, abstract:S5, abstract:S6)
• They also claim the framework provides an efficient and robust alternative with rapid inference and accurate estimation under stochastic volatility. (abstract:S7)

Data, method, or discussion scope

Scope includes the modeling challenge, proposed method, and claimed performance benefits; it does not expose error metrics, mesh/discretization settings, or cross-asset validation details. (abstract:S1, abstract:S2, abstract:S5, abstract:S6, abstract:S7)

Main limitations

The efficiency and robustness wording is abstract-level without explicit convergence rates, error bounds, or failure behavior under boundary-condition conflicts. (abstract:S7, abstract:S6)

Option market making with hedging-induced market impact

• Published: 2026-04-29
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The paper models option market making with hedging-induced underlying price impact, making the interaction of quoting, hedging and inventory explicit, which is critical for feedback control and liquidity risk modeling. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S6)

Main author claims

• The authors claim option market-making is modeled with market-maker hedging impacting the underlying asset, while option order flow is modeled by Cox processes with state- and quote-dependent intensities. (abstract:S1, abstract:S2)
• They analyze feedback-induced manipulation/arbitrage possibilities, formulate a mixed control problem with continuous quoting and impulse hedging, and implement policy-optimization numerics to study interactions among liquidity, inventory risk, and impact. (abstract:S4, abstract:S5, abstract:S6)

Data, method, or discussion scope

Scope includes the demand-process specification, coupled dynamics, manipulation/arbitrage discussion, and a policy-optimization numerical method, but omits parameter estimates, stability conditions, and full comparative performance tables. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main limitations

The abstract emphasizes theoretical design and well-posedness but lacks market calibration data and control-constraint details needed to determine usable risk parameters. (abstract:S5, abstract:S6)

End-to-End Large Portfolio Optimization for Variance Minimization with Neural Networks through Covariance Cleaning

• Published: 2026-04-21
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This work combines covariance cleaning and minimum-variance optimization in a rotation-invariant network, which could reduce retraining burden in large universes; however, claims of broad generalization and superiority require stress-aware verification for deployment. (abstract:S1, abstract:S2, abstract:S3, abstract:S5, abstract:S7)

Main author claims

• The authors claim a rotation-invariant neural network jointly learns lag transformations and covariance eigenvalue regularization for global minimum-variance portfolio construction. (abstract:S1)
• They claim the architecture mirrors the analytical solution form while being dimension-agnostic, scaling from hundreds to around one thousand U.S. equities without retraining, with out-of-sample improvements in realized volatility, maximum drawdown, and Sharpe ratio versus competitors. (abstract:S3, abstract:S5)
• The authors report that the learned covariance can be used in long-only optimizers with virtually no loss of relative advantage, and that the advantage persists with auction execution, empirical slippage, exchange fees, leverage financing, and acute stress episodes. (abstract:S6, abstract:S7)

Data, method, or discussion scope

Scope includes the proposed architecture, claimed dimensional generalization, out-of-sample date span from Jan 2000 to Dec 2024, competitor comparisons including nonlinear shrinkage, and implementation realism with auction execution, slippage, fees, and financing charges. (abstract:S1, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)

Main limitations

Claims of no retraining and persistent advantages omit retraining cadence, drift controls, and implementation details for constrained formulations under changing regimes. (abstract:S3, abstract:S6, abstract:S7)

Realised Volatility Forecasting: Machine Learning via Financial Word Embedding

• Published: 2026-04-13
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

This abstract examines whether news text can improve realised-volatility forecasting when combined with standard benchmarks, which matters for incorporating unstructured signals into risk pipelines and deciding when such signals are operationally useful. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main author claims

• The authors claim they build news-embedding representations for realised-volatility forecasting, using them as a standalone model and as a complement to standard realised-volatility benchmarks. (abstract:S1, abstract:S2)
• They further claim that out-of-sample cross-sectional tests show stronger predictive effects for stock-specific news and high-volatility days, and combining the news signal with a leading benchmark improves statistical performance and economic gains. (abstract:S3, abstract:S4)

Data, method, or discussion scope

The scope includes model framing, embedding-based forecasting design, benchmark comparisons, out-of-sample findings, and explainability framing; it does not provide sample windows, exact metric definitions, or detailed preprocessing steps. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main limitations

The claims rely on abstract-level summaries without noise handling, event-window definitions, or reproducible hyperparameters, preventing direct use in signal-governance decisions. (abstract:S2, abstract:S3, abstract:S4)

Do Prediction Markets Forecast Cryptocurrency Volatility? Evidence from Kalshi Macro Contracts

• Published: 2026-04-01
• Source: arXiv Quantitative Finance History
• Publication status: preprint
• Original source: Open original source

Why it matters

The study links prediction-market probability movements to realized crypto volatility and documents multi-channel and multiple-testing-aware evidence, which is operationally important for macro-to-crypto risk indicators, with external validity bounded by sample and correction design. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)

Main author claims

• The authors claim changes in Kalshi macro prediction-market probabilities forecast crypto realized volatility through separate monetary-policy and recession-risk channels. (abstract:S1, abstract:S2, abstract:S3)
• They report statistical characterization: KXFED with t=3.63, p<0.001 and regime dependence, KXRECSSNBER OOS MSFE ratio 0.979, and significant CPI-channel effects for several altcoins with reported t-statistics and MSFE gains. (abstract:S2, abstract:S3, abstract:S4, abstract:S5)
• The authors report that the Bitcoin-Fed-dovish and Chainlink-CPI results survive Benjamini-Hochberg correction at q=0.05 and retain incremental information after comparisons with Fed Funds futures, Treasury yields, and the Deribit implied-volatility index. (abstract:S5, abstract:S6)

Data, method, or discussion scope

Scope includes sample coverage, reported t-statistics, MSFE ratios, BH correction at q=0.05, and orthogonalization results against conventional instruments, but not the full regression design or full table of covariates. (abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6, abstract:S7)

Main limitations

Although significance is reported, the abstract omits rolling split protocol, drawdown risk controls, trading frictions, and sensitivity to event-window definitions. (abstract:S3, abstract:S4, abstract:S7)