Skip to content

2023 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: 2023-01-01 to 2023-03-31
  • Passed rule review: 8
  • Sources: 7

Topic distribution

Domains

  • volatility: 6
  • execution costs: 3
  • option returns: 2
  • hedging exposure risk: 2
  • microstructure: 2
  • institutional execution: 1

Methods

  • financial ml: 3
  • research methods: 3

Facets

  • instrument vix options: 1
  • instrument single stock options: 1
  • horizon 0dte: 1
  • structure straddle: 1

Passed rule review

Volatility Forecasting with Machine Learning and Intraday Commonality

  • Published: 2023-03-20
  • Source: Journal of Financial Econometrics
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

This peer-reviewed version shares the same core claims as the earlier preprint and thus is structurally more usable as a benchmark; still, universality and superiority claims remain boundary-sensitive. (abstract:S1, abstract:S2, abstract:S3, abstract:S5)

Main author claims

  • The authors forecast intraday realized volatility using pooled intraday commonality and a market-volatility proxy. (abstract:S1)
  • They claim neural networks outperform linear and tree models and remain robust on stocks not included in training. (abstract:S2, abstract:S3)
  • They claim proposed forecasts using past intraday RV and time-of-day effects outperform strong daily-RV-only baselines out of sample. (abstract:S4, abstract:S5)

Data, method, or discussion scope

Evidence is confined to abstract claims about model choice and performance, without explicit forecast metric definitions, error distributions, or hyperparameters. (abstract:S1, abstract:S2, abstract:S5)

Main limitations

No sensitivity analysis is provided for asset-pool size, intraday partition choices, or non-stationary regime-induced performance degradation. (abstract:S3, abstract:S4, abstract:S5)

WTF?! Will 0DTE Cause Gammageddon? With Mike Green and Craig Peterson

  • Published: 2023-03-09
  • Source: The Derivative by RCM Alternatives
  • Publication status: unknown
  • Original source: Open original source

Why it matters

The episode is context-setting around 0DTE and gamma risk, useful for risk-awareness framing, but as synopsis material it is not a standalone empirical source. (description:S5, description:S8, description:S9, description:S13)

Main author claims

  • The authors report: The episode preview says it will cover what 0DTE options are, who trades them, and potential market implications including VIX. (description:S5, description:S13)
  • The authors report: It includes discussions on term choice (quarterly, weekly, daily), gamma hedging, similarity of call/put selling, and potential gamma-explosion risks. (description:S7, description:S8, description:S9)
  • The authors report: The episode includes informational-only and no trade-recommendation disclosures. (description:S17, description:S19, description:S20, description:S22)

Data, method, or discussion scope

Evidence scope is limited to episode outline and chapter topics, with no sample, tests, or market metrics provided. (description:S5, description:S12, description:S13, description:S15)

Main limitations

No reproducible risk parameters, participation-share changes, or event-risk metrics can be inferred from this text, and promotional framing is explicit. (description:S5, description:S10, description:S16, description:S9)

High‐frequency trading and market quality: Evidence from account‐level futures data

  • Published: 2023-02-27
  • Source: Journal of Futures Markets
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

The abstract links account-level HFT participation to market quality and highlights both beneficial and offsetting effects, which is important for market-quality governance despite implementation complexity. (abstract:S1, abstract:S2, abstract:S3, abstract:S4)

Main author claims

  • The authors use intraday regulatory transactions and end-of-day positions to study how HFT participation affects futures market quality. (abstract:S1)
  • They claim higher HFT participation is associated with improved spread and lower Amihud price impact, while aggressive directional position-reduction trading partially offsets these effects. (abstract:S2)
  • They state they exploit the 2015 CME settlement-method change for agricultural commodities in a fixed-effects difference-in-difference design. (abstract:S4, abstract:S5)

Data, method, or discussion scope

Evidence scope is limited to abstract-level panel claims on futures market quality and methods; no sample counts, parameter values, or effect magnitudes are disclosed. (abstract:S1, abstract:S2, abstract:S4, abstract:S5)

Main limitations

The DID identification depends on assumptions (for example parallel trends and unobserved heterogeneity control) that are not inspectable from the abstract alone. (abstract:S4, abstract:S5, abstract:S2)

The Only Constant Is Change: Nonconstant Volatility and Implied Volatility Spreads

  • Published: 2023-02-27
  • Source: Journal of Financial and Quantitative Analysis
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

The paper links implied-volatility spread predictability to aggregate volatility-risk factors, which matters for attribution claims; asserting certain drivers are ruled out is a high-stakes statement at abstract level. (abstract:S1, abstract:S2, abstract:S3, abstract:S5)

Main author claims

  • The authors study predictability of implied-volatility spreads from individual options and argue that VS can arise under simple no-arbitrage conditions when volatility is time-varying. (abstract:S1, abstract:S2)
  • They report that predictability varies systematically with aggregate volatility and is positively related to firms' sensitivity to volatility risk. (abstract:S3)
  • They claim VS-hedge portfolio alpha is explained by aggregate volatility risk factors and cannot be explained by firm-specific informed trading, transaction costs, or liquidity. (abstract:S4, abstract:S5)

Data, method, or discussion scope

The available evidence is abstract-level attribution and exclusion claims, without disclosed definitions for volatility sensitivity, model design, or inferential statistics. (abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main limitations

Claims ruling out informed-trading or liquidity explanations require strict measurement design; the abstract omits counterfactual identification mechanics and liquidity stratification details. (abstract:S2, abstract:S5, abstract:S3)

Option Return Predictability with Machine Learning and Big Data

  • Published: 2023-02-24
  • Source: The Review of Financial Studies
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

The study combines large-scale cross-sectional equity-option prediction, time-separated validation, and transaction-cost sensitivity in one design, allowing comparison of linear and nonlinear models while separating statistical forecast gains from implementable economic returns. (full_text:S24, full_text:S27, full_text:S35, full_text:S39, full_text:S53, full_text:S66)

Main author claims

  • The authors report more than 12 million option-month observations from 1996-2020, using 80 option and 193 stock characteristics to predict next-month delta-hedged excess returns; training, validation, and testing are time-separated and rolled across 18 out-of-sample years from 2003 through 2020. (full_text:S27, full_text:S28, full_text:S29, full_text:S30, full_text:S31, full_text:S187, full_text:S188, full_text:S189, full_text:S190, full_text:S191)
  • The authors report positive out-of-sample R² for every nonlinear model over the full test period, with GBR and Dart at 2.26% and 1.96%; the nonlinear ensemble exceeds the linear ensemble by more than 1.7 percentage points of out-of-sample R². (full_text:S42, full_text:S43, full_text:S44, full_text:S45, full_text:S50)
  • The authors report a 2.04% monthly gross return for the nonlinear forecast-sorted long-short portfolio and 0.67% per month when both option trades and delta hedges pay 100% of quoted spreads; significance disappears only in that most severe cost case. (full_text:S53, full_text:S54, full_text:S55, full_text:S66, full_text:S67, full_text:S68, full_text:S69, full_text:S71, full_text:S72)
  • The authors find predictability increasing with information-friction and composite-mispricing scores, while explicitly acknowledging that these proxies also load on volatility, jump, and liquidity risks and therefore do not rule out risk compensation. (full_text:S781, full_text:S782, full_text:S783, full_text:S826, full_text:S827, full_text:S828, full_text:S833, full_text:S834, full_text:S838)

Data, method, or discussion scope

The evidence covers monthly delta-hedged excess returns on U.S. single-stock calls and puts, expanding-window train/validation/test splits, multiple linear and nonlinear models, cross-sectional forecast metrics, forecast-sorted portfolios, and cost and leverage sensitivities. These are author-reported historical out-of-sample backtests, not independent replication or live execution by this site. (full_text:S27, full_text:S28, full_text:S29, full_text:S31, full_text:S175, full_text:S176, full_text:S180, full_text:S186, full_text:S187, full_text:S188, full_text:S189, full_text:S190, full_text:S191, full_text:S724, full_text:S725)

Main limitations

OptionMetrics does not observe actual trading costs, so the paper proxies effective spreads with 25%-100% of quoted spreads and separately estimates delta-hedging costs; the 0.67% figure is therefore not realized post-trade net performance. Models are fixed annually and estimated on historical U.S. equity options, limiting transfer to other markets, live capacity, and market impact; the friction and mispricing results do not identify a causal mechanism or exclude risk compensation. (full_text:S66, full_text:S67, full_text:S68, full_text:S69, full_text:S187, full_text:S191, full_text:S834, full_text:S838)

Limits of Arbitrage and Primary Risk-Taking in Derivative Securities

  • Published: 2023-02-14
  • Source: The Review of Asset Pricing Studies
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

The abstract frames the tension between theoretical dynamic delta hedging and residual risk under arbitrage limits, which matters for any implementation that treats options as near-redundant while expecting predictable hedged return structure. (abstract:S1, abstract:S2, abstract:S4)

Main author claims

  • The authors state that classical option valuation treats a derivative as redundant relative to the underlying through dynamic delta hedging. (abstract:S1)
  • They argue that although dynamic delta hedging is effective in practice, remaining risk is still large because of practical arbitrage limits. (abstract:S2)
  • The paper claims to quantify percentage variance reduction for US stock-option delta hedges and build a top-down attribution framework for remaining risk sources. (abstract:S4)

Data, method, or discussion scope

Evidence scope is abstract-level theoretical framing and claimed framework components, without sample window, estimation specification, error metric, or significance levels. (abstract:S1, abstract:S2, abstract:S4)

Main limitations

Because this is abstract-only, it does not allow verification across market states, maturity buckets, or operational execution frictions, including transaction-cost effects. (abstract:S2, abstract:S4)

Score-Driven Modeling with Jumps: An Application to S&P500 Returns and Options

  • Published: 2023-02-08
  • Source: Journal of Financial Econometrics
  • Publication status: peer_reviewed
  • Original source: Open original source

Why it matters

A score-driven jump-and-volatility framework with theoretical stationarity/ergodicity conditions is relevant to model consistency, but “excellent fit” needs explicit metrics and robustness context before operational reliance. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main author claims

  • They introduce a score-driven model with two shock sources allowing time-varying volatility and jumps, including time-varying jump intensity. (abstract:S1, abstract:S3)
  • The authors report: In-sample and out-of-sample S&P500 analyses are said to match observed returns and outperform standard GARCH-with-jumps specifications. (abstract:S2, abstract:S4)
  • They also apply it to option pricing via risk-neutralization, claiming reliable implied-volatility surfaces. (abstract:S5, abstract:S6)

Data, method, or discussion scope

Evidence is confined to abstract-level statements on theory, stationarity conditions, and S&P500 demonstration with unquantified “excellent agreement” and outperformance claims. (abstract:S1, abstract:S2, abstract:S4, abstract:S5, abstract:S7)

Main limitations

The abstract omits estimation specifics, risk-neutral mapping assumptions, and reproducibility details; it only flags that supplementary materials exist. (abstract:S2, abstract:S4, abstract:S5, abstract:S7)

Stochastic volatility modeling of high-frequency CSI 300 index and dynamic jump prediction driven by machine learning

  • Published: 2023-01-03
  • Source: arXiv Quantitative Finance History
  • Publication status: preprint
  • Original source: Open original source

Why it matters

The paper targets high-frequency CSI 300 dynamics with ML/DL for faster valuation in settings with asynchrony and long-memory issues; however acceleration and stakeholder suitability claims need explicit operational boundaries. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5, abstract:S6)

Main author claims

  • They adapt a generalized Barndorff-Nielsen and Shephard model to address information asynchrony and long-term dependence issues in CSI300 high-frequency prices. (abstract:S1, abstract:S2)
  • They claim ML/DL algorithms are used to speed valuation and assess forecast outputs. (abstract:S3)
  • They claim jump tracking across magnitudes supports dynamic process simulation and jump prediction, and suggest findings may be useful for investors and regulators. (abstract:S4, abstract:S6)

Data, method, or discussion scope

Evidence scope is limited to model claims and result summaries; it does not include estimator accuracy, noise handling parameters, jump-threshold definitions, or regulatory deployment criteria. (abstract:S1, abstract:S2, abstract:S3, abstract:S4, abstract:S5)

Main limitations

Robustness across market structures and data-quality regimes is not reported; it is unclear whether captured deterministic components imply statistical significance or structural interpretability. (abstract:S5, abstract:S2, abstract:S6)