Paper: arXiv 2410.19751

Authors: Aubain Nzokem, Daniel Maposa

Abstract

The paper proposes and implements a methodology to fit a seven-parameter Generalized Tempered Stable (GTS) distribution to financial data. The nonexistence of the mathematical expression of the GTS probability density function makes the maximum likelihood estimation (MLE) inadequate for providing parameter estimations. Based on the function characteristic and the fractional Fourier transform (FRFT), we provide a comprehensive approach to circumvent the problem and yield a good parameter estimation of the GTS probability. The methodology was applied to fit two heavily tailed data (Bitcoin and Ethereum returns) and two peaked data (S&P 500 and SPY ETF returns). For each index, the estimation results show that the six-parameter estimations are statistically significant except for the local parameter, \(μ\). The goodness-of-fit was assessed through Kolmogorov-Smirnov, Anderson-Darling, and Pearson’s chi-squared statistics. While the two-parameter geometric Brownian motion (GBM) hypothesis is always rejected, the GTS distribution fits significantly with a very high p-value; and outperforms the Kobol, Carr-Geman-Madan-Yor, and Bilateral Gamma distributions.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper is highly theoretical with dense mathematics, featuring complex Lévy measure expressions, characteristic exponents, and cumulative functions derived via fractional Fourier transforms, but it also includes concrete backtests on financial data (Bitcoin, S&P 500) and standard goodness-of-fit metrics, making it mathematically advanced yet empirically grounded.

Research Flowchart

  flowchart TD
  A["Research Goal: Fit GTS distribution<br>to financial data"] --> B["Data Collection<br>Bitcoin, Ethereum, S&P 500, SPY Returns"]
  B --> C["Methodology: MLE & Fractional Fourier Transform"]
  C --> D{"Parameter Estimation"}
  D -->|"Success"| E["Statistical Testing<br>K-S, A-D, Chi-squared"]
  D -->|"Failure"| F["Refine Parameters"]
  F --> C
  E --> G["Key Findings"]
  G --> H["GBM Hypothesis Rejected"]
  G --> I["GTS Outperforms<br>Kobol, CGMY, Bilateral Gamma"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes and implements a methodology to fit a seven-parameter Generalized Tempered Stable (GTS) distribution to financial data.

Source sentence
The paper proposes and implements a methodology to fit a seven-parameter Generalized Tempered Stable (GTS) distribution to financial data.

Assumes 2

  • Observations are independent and identically distributed with a density satisfying regularity conditions. statistical
    Source sentence
    Let X1, ..., Xm be independent and identically distributed (i.i.d) random variables with density f(x|θ) satisfying some regularity conditions
  • The likelihood function can be maximized using Newton‑Raphson iteration assuming a local solution exists. model
    Source sentence
    The solutions, V , in (22) are provided by the Newton-Raphson iteration algorithm formula (23): V n+1 = V n −(I′′ m(x, V n))−1I′ m(x, V n).

Data

  • Bitcoin daily returns crypto-onchain unknown
    Source sentence
    two heavily tailed data (Bitcoin and Ethereum returns)
  • Ethereum daily returns crypto-onchain unknown
    Source sentence
    two heavily tailed data (Bitcoin and Ethereum returns)
  • S&P 500 index prices · Yahoo Finance market-prices public
    04 January 2010 to 22 July 2024 · S&P 500 constituents
    Source sentence
    The data were extracted from Yahoo Finance. The historical prices span from 04 January 2010 to 22 July 2024
  • SPY ETF prices · Yahoo Finance market-prices public
    04 January 2010 to 22 July 2024 · SPY ETF
    Source sentence
    The data were extracted from Yahoo Finance. The historical prices span from 04 January 2010 to 22 July 2024

Recognised sources: CoinMarketCap Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • The GTS distribution fits the four historical data sets significantly better than the two‑parameter GBM, which is always rejected. statistical
    Source sentence
    While the two-parameter GBM hypothesis is always rejected, the GTS hypothesis is accepted and yields a high p-value.
  • Log‑likelihood, AIC and BIC statistics show that the seven‑parameter GTS outperforms the Normal (GBM) model for each index. statistical
    Source sentence
    The log-likelihood, Akaike’s information Criteria (AIC), and Bayesian information criteria (BIK) statistics show that the GTS distribution with seven parameters performs better than the two-parameter Normal distribution (GBM).
  • For Bitcoin and Ethereum, the location parameter µ is negative and not statistically significant at the 5% level, while other parameters are positive and significant. statistical
    Source sentence
    The asymptotic standard error for µ is quite large and suggests that µ is not statistically significant at 5%.
  • Skewness parameters (λ+, λ−) for Bitcoin are statistically significant at 5%, indicating asymmetry and left‑skewness. statistical
    Source sentence
    The skewness parameters (λ+, λ−) are statistically significant at 5%.
  • The GTS distribution outperforms the Bilateral Gamma distribution for the S&P 500 and SPY ETF indexes according to p‑values. statistical
    Source sentence
    the GTS distribution outperforms the Bilateral Gamma distribution for the S&P 500 and SPY ETF indexes.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 24 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (57)

  1. An introduction to probability theory and its applications, Volume 2 (1971)
  2. Precision and Accuracy (2007) — link
  3. On the Combinatorics of Cumulants (2000) — link
  4. Fourier transform methods in finance (2010)
  5. Fourier-Based Valuation Methods in Mathematical Finance (2014) — link
  6. Analysis of Fourier transform valuation formulas and applications (2010)
  7. Likelihood ratio tests for model selection and non-nested hypotheses (1989)
  8. Relations between cumulants and central moments and their applications (2021) — link
and 49 more
  1. Parametric Estimation of Tempered Stable Laws (2024) — link
  2. Simulating stock prices using geometric Brownian motion model under normal and convoluted distributional assumptions (2023) — link
  3. L 'e vy processes and infinitely divisible distributions (1999)
  4. Modeling with Stable Distributions (2020) — link
  5. A Recursive Formulation of the Old Problem of Obtaining Moments from Cumulants and Vice Versa (1995)
  6. Bitcoin: A peer-to-peer electronic cash system
  7. Financial volatility, L 'e vy processes and power variation
  8. Bilateral gamma distributions and processes in financial mathematics (2008) — link
  9. A new tempered stable distribution and its application to finance (2009)
  10. Stable and Tempered Stable Distributions (2011) — link
  11. Stochastic volatility for L 'e vy processes (2003)
  12. Non-Gaussian Merton-Black-Scholes Theory (2002)
  13. Gamma variance model: Fractional Fourier Transform ( FRFT ) (2021) — link
  14. Modelling tail risk with tempered stable distributions: an overview (2021)
  15. Quantile-based inference for tempered stable distributions (2019)
  16. Handbook of Heavy-Tailed Distributions in Asset Management and Risk Management (2019) — link
  17. Basic Results on L 'e vy Processes (2001) — link
  18. Elements of large-sample theory (1999)
  19. A Course in the Large Sample Theory of Statistical Inference (2023)
  20. Statistical theory and inference (2014)
  21. Statistical Inference (2024) — link
  22. Anderson–Darling Test (2008) — link
  23. Anderson–Darling Tests of Goodness-of-Fit (2011) — link
  24. Sulla determinazione empirica di una legge didistribuzione (1933)
  25. Evaluating the anderson-darling distribution (2004)
  26. EDF statistics for goodness of fit and some comparisons (1974)
  27. Distribution of the Anderson-Darling statistic (1961)
  28. Pricing European Options under Stochastic Volatility Models: Case of Five-Parameter Variance-Gamma Process (2023) — link
  29. Lévy distributions (1997) — link
  30. Stable Distributions (2005) — [link](https://doi.org/10.1007/3-540-27395-6\textunderscore 1)
  31. Tempered stable distributions and processes (2013)
  32. Do financial returns have finite or infinite variance? A paradox and an explanation (2010)
  33. Fitting Generalized Tempered Stable distribution: Fractional Fourier Transform (FRFT) Approach (2022) — link
  34. Enhanced the Fast Fractional Fourier Transform (FRFT) scheme using the closed Newton-Cotes rules (2023) — link
  35. The Ornstein–Uhlenbeck Process and Variance Gamma Process: Parameter Estimation and Simulations (2023) — link
  36. Self-Decomposable Laws Associated with General Tempered Stable (GTS) Distribution and their Simulation Applications (2024) — link
  37. Bitcoin versus S & P 500 Index: Return and Risk Analysis — → 2310.02436
  38. European Option Pricing Under Generalized Tempered Stable Process: Empirical Analysis (2023) — link
  39. Numerical solution of a Gamma - integral equation using a higher order composite Newton-Cotes formulas (2021) — link
  40. The advanced theory of statistics (1945)
  41. The variance gamma process and option pricing (1998)
  42. L 'e vy processes in finance: pricing financial derivatives (2003)
  43. The Kolmogorov-Smirnov test for goodness of fit (1951)
  44. Evaluating Kolmogorov’s distribution (2003)
  45. Computing the Kolmogorov-Smirnov distribution when the underlying CDF is purely discrete, mixed, or continuous (2020)
  46. Rachunek prawdopodobie 'n stwa i statystyka matematyczna w zadaniach (1999)
  47. Applied statistical inference (2014)
  48. Fitting Infinitely divisible distribution: Case of Gamma-Variance Model (2021) — link
  49. Wiley Series in Computational Statistics (2013)

1 of 57 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.