Pith. sign in

REVIEW 3 major objections 6 minor 33 references

Latent-Regime Bias Auditing for Volatility Forecasting

T0 review · 3 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Volatility forecasts that look accurate on average can be badly biased in specific market regimes.

desk verdict Useful audit framework held back by an overclaimed 'latent' regime story and under-specified metrics; the empirical core on conditional bias is sound enough to warrant a serious revision. read the letter →

arxiv 2608.01599 v1 pith:7WWJIRHX submitted 2026-08-03 cs.LG

Arthur Chagas , Pedro Bento , Yan Aquino , Arthur Buzelin , Wagner Meira Jr. , Cristiano Arbex Valle This is my paper · ORCID
classification cs.LG
keywords volatilityforecastingforecastevaluationlatentregimesregime-conditionalbiastailunderpredictionasymmetriclosstime-seriesrepresentationlearningasset-relative
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Volatility forecasts are usually judged by average error. This paper argues that such aggregate metrics hide where and when forecasts become unreliable, and proposes a model-agnostic audit that checks forecast errors inside latent market regimes learned separately for each asset. Applying the audit to daily volatility forecasts for cryptocurrencies and ETFs, the paper shows that models with similar RMSE can differ sharply in regime-specific bias, tail underprediction, and underprediction-sensitive losses. The takeaway is that a forecast can look accurate on average while being economically dangerous in stressed states, so evaluation should ask not only 'which model is most accurate' but 'where does accuracy fail'.

What carries the argument

The load-bearing object is the asset-relative aligned regime partition. Learned embeddings of 64-day market-state windows are normalized within each asset and passed through an adversarial encoder that reduces linearly recoverable asset identity; K-means then clusters each asset's training embeddings into three states, and the clusters are ordered by a stress score into calm, intermediate, and stress regimes. The audit statistics RBA_stable and RCRG_norm compare the largest regime bias against the aggregate bias, and tail-weighted asymmetric loss penalizes underprediction in high-volatility outcomes.

What would settle it

Replace the learned embeddings with a direct stress-score quantile partition (e.g., thirds of the training-period stress score) and rerun the audit. If the RBA_stable, tail-underprediction, and economic-loss results are nearly identical, the latent-regime machinery is not adding information over the stress score alone.

Watch

Extended reading notes

Core claim

The paper's central claim is that aggregate accuracy can mask substantial regime-dependent failures in volatility forecasting. It constructs asset-relative regimes — calm, intermediate, and stress — from window embeddings of market features, using within-asset normalization, adversarial reduction of asset identity, K-means clustering on training data, and alignment by a training-period stress score. On held-out data, it then measures regime-conditional bias, tail-underprediction, and asymmetric economic loss for a wide range of forecasters, from HAR/IV baselines to recurrent, convolutional, and Transformer models. The result is that RMSE ranking, conditional-reliability ranking, and economic

Load-bearing premise

The asset-relative regimes are aligned by a training-period stress score; if the learned embeddings add no information beyond a simple stress-score threshold, the central contribution of 'latent' regimes is not substantiated.

Editorial extensions

If this is right

  • A model's rank under RMSE can differ from its rank under regime-conditional reliability and under tail-weighted economic loss.
  • Low aggregate RMSE does not guarantee low worst-regime bias: the audit finds statistically supported conditional bias in at least one regime for several accurate models.
  • Simple finance-native HAR-RV variants can become more attractive under underprediction-sensitive losses than more complex neural or Transformer models.
  • The audit is model-agnostic and out-of-sample: regimes are built only from training information, so any forecaster can be audited without retraining.
  • Choosing a volatility forecaster for risk management should weigh where errors occur, not just average error.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural test of the central contribution would be to run the same audit with a plain stress-score quantile partition in place of the learned embeddings; if the reliability signal is unchanged, the latent representation is adding little.
  • Because regimes are asset-relative, the audit avoids pooled clusters that mainly encode asset identity, suggesting the method could transfer to multi-asset settings where universal regimes are poorly defined.
  • The regime definitions are sample-dependent audit partitions, so extending the audit to longer samples or intraday targets could reveal additional failure modes, but the stability of the latent regimes across such settings is an open question.
  • If regulators or risk managers adopt conditional reliability reporting, model selection pressure would shift toward forecasters that remain calibrated in stress states, which could change which models are deployed.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes a model-agnostic audit framework for volatility forecasts. It learns asset-normalized time-series embeddings with a TS2Vec-style encoder and an adversarial asset classifier, clusters the embeddings within each asset, aligns the resulting clusters by a training-period stress score into calm/intermediate/stress regimes, and then evaluates a diverse set of volatility forecasters conditionally on these regimes. The empirical application uses daily data for BTC, ETH, and five US ETFs, and the main finding is that models with competitive aggregate RMSE can exhibit systematic regime-conditional bias, tail underprediction, and unfavorable economic-loss profiles. The paper also introduces two summary diagnostics, RBA_stable and RCRG_norm, and reports block-bootstrap confidence intervals for regime-specific bias. The central claim is that aggregate accuracy can mask conditionally unreliable forecasts.

Significance. If the framework is taken at face value, it addresses a real gap: standard aggregate metrics such as RMSE and MAE do compress away the location of forecast failures, and for volatility forecasting those failures are most consequential in stressed states. The paper has several strengths: the regimes are constructed on training information only and assigned out of sample, which avoids the most naive forms of leakage; the use of block-bootstrap confidence intervals for regime bias is appropriate given overlapping forecast horizons; the evaluation spans a broad set of model families, including finance-native HAR benchmarks, recurrent, convolutional, and transformer-style architectures; and the repository is provided for reproducibility. The paper also carefully avoids overclaiming that the regimes are universal market states, acknowledging they are sample-dependent audit partitions. However, the central novelty of the paper — the 'latent' regime construction — is not yet substantiated, because the alignment step uses a stress score built from essentially the same variables used to learn the embeddings. A simple stress-score thresholding baseline would clarify whether the learned repres

major comments (3)
  1. [Section III-C and Table VII] The paper's central novelty (contribution 3) is the construction of 'asset-relative aligned regimes' from learned embeddings. However, the alignment step orders clusters by a training-period stress score whose components — volatility, drawdown, absolute return, and stress-proxy variables — are also inputs to the embedding (Section III-A). The final calm/intermediate/stress labels are therefore monotone in a visible economic score, and it is unclear whether the latent representation adds any information beyond a direct thresholding of that score. The paper should add a baseline in which regimes are defined by within-asset quantiles of the training-period stress score (with the same K=3), and then compare the audit conclusions (Table V, Table VI, and Table VII) under that baseline. If the latent regimes give materially different conditional-bias findings or asset-regime dependence, contrib
  2. [Sections III-E and III-F, Tables V and VI] The headline summary statistics RBA_stable and RCRG_norm are defined with hyperparameters that are never reported: lambda_stab, epsilon, and the precise definition of the valid-regime set K_valid. Similarly, the asymmetric loss used for the economic-loss rankings is defined as lambda > 1, but the actual value of lambda used in Table V and Figure 2 is not given. Since the model rankings in Figure 2 depend on these choices, the main empirical claims are not reproducible from the manuscript alone. Please report the exact values and provide a sensitivity analysis over lambda_stab and lambda. The stress score used for alignment is also described only qualitatively ('combining volatility, drawdown, absolute return, and stress-proxy information'); its exact formula and weighting must be specified, both for reproducibility and for the thresholding baseline suggested above.
  3. [Section IV-A and Table V] The empirical comparison is described only at a coarse level. The paper does not give the feature set, training procedure, hyperparameter choices, or validation scheme for each of the sixteen forecasting models. While the repository link is a useful step, the manuscript itself needs at least a table summarizing each model's key configuration (or a reference to a fixed, versioned configuration file) so that the reported RMSE and regime-bias rankings can be independently checked. This is especially important because the paper draws conclusions about model families — e.g., that transformer-style models are less responsive to volatility extremes — which depend on the specific implementations and hyperparameters used.
minor comments (6)
  1. [Table III] The first column header appears garbled as 'RegimenCrypto'. It should be split into separate columns, e.g., 'Regime', 'n', and 'Crypto share', so that the sample sizes and asset composition are legible.
  2. [References [28]] The author list for the domain-adversarial training reference is misprinted as 'Y. Gani, E. Ustinova, ...'; the correct first author is Yaroslav Ganin, and the standard citation is Ganin et al., JMLR 2016.
  3. [Section III-F] The asymmetric loss uses a parameter lambda > 1, but no concrete value is given in the text. The value might be in the repository, but it should appear in the paper because Table V and Figure 2 report ranking under this loss.
  4. [Table VII] The robustness table reports median RBA and RCRG values without uncertainty estimates. Given that the main results rely on bootstrap inference, it would be helpful to provide bootstrap intervals for these medians as well, even if only for the main specification.
  5. [Figure 1] The right panel is labeled 'Economic stress gradient' but does not clearly identify which color scale corresponds to the stress proxy. Adding a color bar with numeric values would improve readability.
  6. [Section IV-A] The term 'historical baselines' is vague. The table later lists 'CurrentHistRV', but the reader is not told exactly how this baseline is constructed (e.g., rolling historical average or last observed RV). Please define it explicitly.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor self-definitional validation of stress-aligned regimes; central audit claim is independent.

  1. self definitional [Section III-C (regime construction) and Section V-A / Table III (RQ1 validation)]
    "Clusters are ordered by average stress score and labeled as ci,t ∈ {calm,intermediate,stress}. ... Table III shows that the aligned regimes differ along interpretable financial dimensions. Calm periods have lower realized and implied volatility, smaller drawdowns, and no tail-event concentration."

    Regime labels are defined by ordering clusters on a training-period stress score that explicitly combines realized volatility, absolute returns, high–low range, drawdown, and stress-proxy variables (Section IV-B). The RQ1 validation then reports that calm regimes have lower RV/drawdown and stress regimes higher; this is entailed by the ordering rule, so Table III is a restatement of the alignment construction, not independent evidence that the learned embeddings capture structure beyond the stress score. The central audit result—that aggregate RMSE can hide regime-conditional bias—is computed from out-of-sample forecast errors and does not reduce to this construction, so the circularity is limited to the 'economic meaningfulness' validation.

full rationale

The main derivation—conditioning out-of-sample forecast errors on within-asset cluster regimes and comparing aggregate vs. regime-conditional bias—is self-contained: regimes are built from market features without using forecast errors, and the reported hidden-bias findings are empirical regularities, not identities. The one definitional coupling is the RQ1 validation: since the calm/intermediate/stress labels are assigned by ordering clusters on a stress score composed of RV, drawdown, absolute returns and stress proxies, Table III's finding that those variables differ across regimes is a restatement of the labeling rule. This does not invalidate the central claim; it only means the 'latent' regimes are not shown to add information beyond the visible stress score. No self-citation is load-bearing, and no forecast is fitted to the regimes. Overall score 2.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

The framework does not introduce physical entities. Its construction rests on several domain assumptions about the sufficiency of features, the validity of realized volatility, and the cross-asset comparability of stress-aligned regimes.

free parameters (7)
  • L (window length) = 64
    Chosen for the main specification; not justified or swept.
  • d (embedding dimension) = 64
    Chosen for the main specification; not justified or swept.
  • K (number of regimes) = 3
    Chosen; other K values not tested.
  • lambda (asymmetric loss weight) = not reported
    In the economic-loss metric, lambda > 1 but value not given.
  • lambda_stab (denominator weight in RBA_stable) = not reported
    Affects rankings; value not stated in text.
  • epsilon (stability constant) = not reported
    Prevents division by zero; value not stated.
  • block length for bootstrap = 10 days
    Chosen; sensitivity not tested.
assumptions (4)
  • domain assumption Realized volatility from squared returns is an unbiased proxy for latent volatility
    Standard in the literature; the target y is defined this way.
  • domain assumption The feature set (returns, range, IV, volume, drawdown, stress) is sufficient to characterize market state relevant to forecast errors
    The regime construction and stress alignment rely entirely on these features.
  • domain assumption Assigning held-out windows to the nearest training centroid yields stable, meaningful regimes
    Assumes distribution shift does not invalidate the out-of-sample regime assignment.
  • domain assumption The stress-score ordering makes regimes comparable across assets
    Without this, cluster labels would be arbitrary across assets.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Latent-Regime Bias Auditing for Volatility Forecasting." pith.science (2026). https://pith.science/paper/7WWJIRHX

@misc{pith2026260801599,
  author       = {Pith},
  title        = {Pith review of: Latent-Regime Bias Auditing for Volatility Forecasting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7WWJIRHX}},
  note         = {Machine review of arXiv:2608.01599}
}
read the original abstract

Volatility forecasts are commonly evaluated with aggregate accuracy metrics such as RMSE and MAE, but these metrics can hide conditional failures that matter for risk management. This paper proposes a model-agnostic audit framework for evaluating whether volatility forecasts remain reliable across latent market regimes. We learn time-series representations of market-state windows, cluster them into regimes using only training information, assign regimes out of sample, and compare aggregate forecast behavior with regime-conditional bias, tail-underprediction, and underprediction-sensitive economic losses. Applied to daily volatility forecasting across cryptocurrency and ETF assets, the audit shows that models with competitive aggregate accuracy can still exhibit substantial regime-specific bias and severe tail underprediction. The results suggest that volatility forecasting should be evaluated not only by average error, but also by where and how forecasts become unreliable. Our framework shifts forecast evaluation from asking which model is most accurate on average to identifying the market regimes in which apparently accurate forecasts fail conditionally. Reproducibility: https://github.com/arthurchagas1/Latent-Regime-Bias-Auditing-for-Volatility-Forecasting

Figures

Figures reproduced from arXiv: 2608.01599 by the authors.

Figure 1
Figure 1. Asset-relative latent geometry after adversarial representation learning. The left panel shows the economically aligned asset-relative [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Metric-dependent model rankings. The left panel compares each model’s aggregate RMSE rank with its conditional-reliability rank, [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Hidden bias under asset-relative aligned regimes. Gray bars [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 26 canonical work pages

  1. [1]

    Evaluating interval forecasts,

    P. F. Christoffersen, “Evaluating interval forecasts,”International Economic Review, vol. 39, no. 4, pp. 841–862, 1998. [Online]. Available: http://www.jstor.org/stable/2527341

    work page arXiv 1998

  2. [2]

    Estimation and testing of forecast rationality under flexible loss,

    G. Elliott, I. Komunjer, and A. Timmermann, “Estimation and testing of forecast rationality under flexible loss,”The Review of Economic Studies, vol. 72, no. 4, pp. 1107–1125, 2005. [Online]. Available: http://www.jstor.org/stable/3700702

    work page arXiv 2005

  3. [3]

    A new approach to the economic analysis of nonstationary time series and the business cycle,

    J. D. Hamilton, “A new approach to the economic analysis of nonstationary time series and the business cycle,”Econometrica, vol. 57, no. 2, pp. 357–384, 1989. [Online]. Available: http://www.jstor.org/stable/1912559

    arXiv 1989

  4. [4]

    Tests of conditional predictive ability,

    R. Giacomini and H. White, “Tests of conditional predictive ability,”Econometrica, vol. 74, no. 6, pp. 1545–1578, 2006. [Online]. Available: http://www.jstor.org/stable/4123083

    work page arXiv 2006

  5. [5]

    Multicalibration: Calibration for the (Computationally- identifiable) masses,

    U. Hebert-Johnson, M. Kim, O. Reingold, and G. Rothblum, “Multicalibration: Calibration for the (Computationally- identifiable) masses,” inProceedings of the 35th International Conference on Machine Learning, ser. Proceedings of Machine Learning Research, J. Dy and A. Krause, Eds., vol. 80. PMLR, 10–15 Jul 2018, pp. 1939–1948. [Online]. Available: https:/...

    work page 2018

  6. [6]

    Autoregressive conditional heteroscedasticity with estimates of the variance of united kingdom inflation,

    R. F. Engle, “Autoregressive conditional heteroscedasticity with estimates of the variance of united kingdom inflation,” Econometrica, vol. 50, no. 4, pp. 987–1007, 1982. [Online]. Available: http://www.jstor.org/stable/1912773

    arXiv 1982

  7. [7]

    Generalized autoregressive conditional heteroskedasticity,

    T. Bollerslev, “Generalized autoregressive conditional heteroskedasticity,”Journal of Econometrics, vol. 31, no. 3, pp. 307–327, 1986. [Online]. Available: https: //www.sciencedirect.com/science/article/pii/0304407686900631

    work page arXiv 1986

  8. [8]

    Modeling and forecasting realized volatility,

    T. G. Andersen, T. Bollerslev, F. X. Diebold, and P. Labys, “Modeling and forecasting realized volatility,”Econometrica, vol. 71, no. 2, pp. 579–625, 2003. [Online]. Available: http://www.jstor.org/stable/3082068

    work page arXiv 2003

Show all 33 references
  1. [9]

    Roughing it up: Including jump components in the measurement, modeling, and forecasting of return volatility,

    T. G. Andersen, T. Bollerslev, and F. X. Diebold, “Roughing it up: Including jump components in the measurement, modeling, and forecasting of return volatility,”The Review of Economics and Statistics, vol. 89, no. 4, pp. 701–720, 2007. [Online]. Available: http://www.jstor.org...

    2007

  2. [10]

    A simple approximate long-memory model of realized volatility,

    F. Corsi, “A simple approximate long-memory model of realized volatility,”Journal of Financial Econometrics, vol. 7, no. 2, pp. 174–196, 03 2009. [Online]. Available: https: //doi.org/10.1093/jjfinec/nbp001

    2009 doi

  3. [11]

    The relation between implied and realized volatility,

    B. Christensen and N. Prabhala, “The relation between implied and realized volatility,”Journal of Financial Economics, vol. 50, no. 2, pp. 125–150, 1998. [Online]. Available: https://www. sciencedirect.com/science/article/pii/S0304405X98000348

    1998

  4. [12]

    The role of implied volatility in forecasting future realized volatility and jumps in foreign exchange, stock, and bond markets,

    T. Busch, B. Christensen, and M. Nielsen, “The role of implied volatility in forecasting future realized volatility and jumps in foreign exchange, stock, and bond markets,” 01 2007

    2007

  5. [13]

    V olatility forecast comparison using imperfect volatility proxies,

    A. J. Patton, “V olatility forecast comparison using imperfect volatility proxies,”Journal of Econometrics, vol. 160, no. 1, pp. 246–256, 2011, realized V olatility. [Online]. Available: https:// www.sciencedirect.com/science/article/pii/S030440761000076X

    2011

  6. [14]

    Novel volatility forecasting using deep learning–long short term memory recurrent neural networks,

    Y . Liu, “Novel volatility forecasting using deep learning–long short term memory recurrent neural networks,”Expert Syst. Appl., vol. 132, no. C, p. 99–109, Oct. 2019. [Online]. Available: https://doi.org/10.1016/j.eswa.2019.04.038

    2019 doi

  7. [15]

    V olatility forecasting with machine learning and intraday commonality*,

    C. Zhang, Y . Zhang, M. Cucuringu, and Z. Qian, “V olatility forecasting with machine learning and intraday commonality*,” Journal of Financial Econometrics, vol. 22, no. 2, pp. 492–530, 06 2024. [Online]. Available: https://doi.org/10.1093/jjfinec/ nbad005

    2024 doi

  8. [16]

    Garch-informed neural networks for volatility prediction in financial markets,

    Z. Xu, J. Liechty, S. Benthall, N. Skar-Gislinge, and C. McComb, “Garch-informed neural networks for volatility prediction in financial markets,” inProceedings of the 5th ACM International Conference on AI in Finance, ser. ICAIF ’24. New York, NY , USA: Association for Computi...

    2024

  9. [17]

    Repurposing language models for fx volatility forecasting: A data-efficient and context- aware approach,

    Q. A. Nguyen, C. Guo, and W. Luk, “Repurposing language models for fx volatility forecasting: A data-efficient and context- aware approach,” inProceedings of the 6th ACM International Conference on AI in Finance, ser. ICAIF ’25. New York, NY , USA: Association for Computing Ma...

    2025

  10. [18]

    Prediction of realized volatility and implied volatility indices using ai and machine learning: A review,

    E. S. Gunnarsson, H. R. Isern, A. Kaloudis, M. Risstad, B. Vigdel, and S. Westgaard, “Prediction of realized volatility and implied volatility indices using ai and machine learning: A review,”International Review of Financial Analysis, vol. 93, p. 103221, 2024. [Online]. Avail...

    2024

  11. [19]

    Ts2vec: Towards universal representation of time series,

    Z. Yue, Y . Wang, J. Duan, T. Yang, C. Huang, Y . Tong, and B. Xu, “Ts2vec: Towards universal representation of time series,” inAAAI Conference on Artificial Intelligence, 2021. [Online]. Available: https://api.semanticscholar.org/CorpusID:237497421

    2021

  12. [20]

    A time series is worth 64 words: Long-term forecasting with transformers,

    Y . Nie, N. H. Nguyen, P. Sinthong, and J. Kalagnanam, “A time series is worth 64 words: Long-term forecasting with transformers,” inThe Eleventh International Conference on Learning Representations, 2023. [Online]. Available: https: //openreview.net/forum?id=Jbdc0vTOcol

    2023

  13. [21]

    itransformer: Inverted transformers are effective for time series forecasting,

    Y . Liu, T. Hu, H. Zhang, H. Wu, S. Wang, L. Ma, and M. Long, “itransformer: Inverted transformers are effective for time series forecasting,” inThe Twelfth International Conference on Learning Representations, 2024. [Online]. Available: https://openreview.net/forum?id=JePfAI8fah

    2024

  14. [22]

    Are transformers effective for time series forecasting?

    A. Zeng, M. Chen, L. Zhang, and Q. Xu, “Are transformers effective for time series forecasting?” in Proceedings of the Thirty-Seventh AAAI Conference on Artificial Intelligence and Thirty-Fifth Conference on Innovative Applications of Artificial Intelligence and Thirteenth Sym...

    2023 doi

  15. [23]

    Unveiling recurring financial patterns: Novel unsupervised filtering algorithms for enhanced forecasting,

    A. Ibrain, V . Hern ´andez, and L. Peinado, “Unveiling recurring financial patterns: Novel unsupervised filtering algorithms for enhanced forecasting,” inProceedings of the 5th ACM International Conference on AI in Finance, ser. ICAIF ’24. New York, NY , USA: Association for C...

    2024

  16. [24]

    Stock embeddings: Learn- ing distributed representations for financial assets,

    R. Dolphin, B. Smyth, and R. Dong, “Stock embeddings: Learn- ing distributed representations for financial assets,”arXiv preprint arXiv:2202.08968, 2022

    2022 arXiv

  17. [25]

    Comparing predictive accuracy,

    F. X. Diebold and R. S. Mariano, “Comparing predictive accuracy,”Journal of Business & Economic Statistics, vol. 13, no. 3, pp. 253–263, 1995. [Online]. Available: http://www.jstor. org/stable/1392185

    1995

  18. [26]

    A reality check for data snooping,

    H. White, “A reality check for data snooping,”Econometrica, vol. 68, no. 5, pp. 1097–1126, 2000. [Online]. Available: http://www.jstor.org/stable/2999444

    2000

  19. [27]

    A test for superior predictive ability,

    P. R. Hansen, “A test for superior predictive ability,”Journal of Business & Economic Statistics, vol. 23, no. 4, pp. 365–380,

  20. [28]

    Domain- adversarial training of neural networks,

    Y . Gani, E. Ustinova, H. Ajakan, P. Germain, H. Larochelle, F. Laviolette, M. Marchand, and V . Lempitsky, “Domain- adversarial training of neural networks,”Journal of Machine Learning Research, 05 2015

    2015

  21. [29]

    Spot API Documentation: Kline/Candlestick Data,

    Binance, “Spot API Documentation: Kline/Candlestick Data,” https://developers.binance.com/docs/binance-spot-api-docs/ rest-api/market-data-endpoints, 2026, accessed: 2026-05-10

    2026

  22. [30]

    yfinance: Download Market Data from Yahoo! Finance’s API,

    R. Aroussi, “yfinance: Download Market Data from Yahoo! Finance’s API,” https://ranaroussi.github.io/yfinance/, 2026, ac- cessed: 2026-05-10

    2026

  23. [31]

    Deribit API Documentation: pub- lic/get volatility index data,

    Deribit, “Deribit API Documentation: pub- lic/get volatility index data,” https://docs.deribit.com/ api-reference/market-data/public-get volatility index data, 2026, accessed: 2026-05-10

    2026

  24. [32]

    CBOE V olatility Index,

    Chicago Board Options Exchange, “CBOE V olatility Index,” Retrieved from FRED, Federal Reserve Bank of St. Louis, 2026, accessed: 2026-05-10. [Online]. Available: https://fred. stlouisfed.org/

    2026

  25. [2005]

    Available: http://www.jstor.org/stable/27638834

    [Online]. Available: http://www.jstor.org/stable/27638834

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.