Fact-checked by Grok 2 weeks ago

Variance-gamma distribution

The variance-gamma distribution, also known as the generalized Laplace distribution or Bessel function distribution, is a four-parameter family of continuous probability distributions defined on the real line, arising as a variance-mean mixture of a normal distribution where the mixing distribution for the variance (and mean) is gamma. It is parameterized by r > 0 (shape of the gamma mixing distribution), \theta \in \mathbb{R} (asymmetry or drift parameter), \sigma > 0 (scale or volatility parameter), and \mu \in \mathbb{R} (location parameter), often denoted VG(r, \theta, \sigma, \mu). The probability density function involves the modified Bessel function of the second kind and takes the form
p(x) = \frac{1}{\sigma \sqrt{\pi} \Gamma(r/2)} \exp\left( \frac{\theta (x - \mu)}{\sigma^2} \right) \left( \frac{|x - \mu|}{2 \sqrt{\theta^2 + \sigma^2}} \right)^{r - 1/2} K_{r - 1/2} \left( \frac{\sqrt{\theta^2 + \sigma^2}}{\sigma^2} |x - \mu| \right),
where K_\nu is the modified Bessel function of the second kind.
Introduced by Madan and Seneta in as a model for share market returns, the distribution captures key empirical features of financial data such as , excess , and heavy tails while possessing finite moments of all orders. Its mean is \mu + r \theta, variance is r (\sigma^2 + 2 \theta^2), and higher moments like and are explicitly computable, enabling flexible modeling of and tail behavior. include the normal distribution (as r \to \infty), (when \theta = 0 and \sigma \to 0), and (symmetric case with r = 2, \theta = 0). The variance-gamma process, a with stationary independent increments and the variance-gamma distribution as its marginals, was further developed by Madan, Carr, and in 1998 for option , generalizing the Black-Scholes model by incorporating jumps via a gamma time-changed with drift. This construction allows the process to exhibit infinite activity (infinitely many jumps in finite time) but finite variation, making it suitable for high-frequency . Applications extend beyond finance to areas like , , and even non-financial data such as particle sizes in or approximations in processes on space. Parameter estimation methods include moment matching, maximum likelihood, and Bayesian approaches, with the distribution's tractable facilitating Fourier-based and .

Introduction

Overview

The variance-gamma distribution is a continuous supported on the entire real line, constructed as a normal variance-mean mixture in which the mixing distribution for the variance follows a . This formulation results in the associated variance-gamma process being an infinite activity , characterized by infinitely many small jumps over any time interval. Key features of the distribution include its capacity for asymmetry and heavier tails relative to the normal distribution, enabling it to model data with pronounced and excess . These properties make it particularly suitable for representing phenomena exhibiting non-Gaussian behavior, such as financial asset returns and turbulent wind speeds. The distribution was developed to address the shortcomings of the normal distribution in capturing the skewed and leptokurtic nature of empirical data in fields like and environmental modeling. It forms a special case within the broader class of generalized hyperbolic distributions, inheriting their flexibility while offering a more tractable structure for certain applications.

Historical Development

The variance-gamma distribution traces its origins to statistical work in the early , where it was first documented in 1929 by , G. B. Jeffery, and Ethel M. Elderton as the exact of the sample derived from a bivariate . In 1932, A. further developed this distribution, naming it the Bessel function distribution due to its integral representation involving modified Bessel functions, and explored its basic properties in the context of . An earlier, less formal appearance occurred in 1973 when H. Sichel applied a related form in the statistical valuation of diamondiferous deposits, though without explicit recognition as a distinct distribution. The probabilistic framework underlying the variance-gamma process emerged from broader research on es during the mid-20th century. The concept of subordination—replacing deterministic time with a random subordinator, such as a , in a —was pioneered by in 1955 as a method to generate new Markov semigroups from existing ones. This technique built on earlier theory from the 1930s and 1940s, including work by Paul Lévy on infinitely divisible distributions, and gained traction in the through studies of processes and subordinators by researchers like and Eugene Lukacs. In , Peter Clark applied subordination in 1973 to explain non-normal asset return distributions by time-changing a Brownian motion with a gamma process, laying groundwork for variance-gamma-like models. The distribution's formal introduction to occurred in 1990, when Dilip B. Madan and Eugene Seneta proposed the symmetric variance-gamma model for share market returns, representing it as a gamma-subordinated without drift to capture excess and clustering of . Building directly on this, Madan, Peter Carr, and Eric C. Chang extended the framework in 1998 to include via a drift term in the subordinated , deriving closed-form option pricing formulas and emphasizing its utility for modeling jumps in asset prices. Madan and Wim J. A. Milne further refined the skewed version in 1991, providing additional probabilistic characterizations. Following these foundational contributions, the variance-gamma model saw widespread adoption in quantitative finance after 2000 for capturing jump dynamics in high-frequency data and improving derivative pricing beyond Black-Scholes limitations. Key milestones include its integration into risk-neutral valuation frameworks and empirical validations, such as those by Seneta in 2004 for fitting financial time series. Extensions to multivariate forms proliferated in the 2010s, with P. Semeraro introducing a multivariate variance-gamma process via subordination with a multivariate gamma subordinator in 2008 for portfolio modeling and option pricing on baskets. Further developments, such as those by E. Luciano and Semeraro in 2010, incorporated dependence structures via generalized gamma convolutions. As of 2025, the distribution is implemented in standard risk management software, including the R package 'VarianceGamma' released in 2018 for simulation and estimation, supporting its routine use in volatility forecasting and stress testing.

Mathematical Definition

Parameters

The variance-gamma distribution is commonly parameterized by four parameters: a location parameter \mu \in \mathbb{R}, a scale parameter \sigma > 0, an asymmetry parameter \theta \in \mathbb{R}, and a shape parameter r > 0. The \mu shifts the entire distribution along the real line without affecting its shape or spread. The \sigma controls the overall dispersion or width of the distribution, with larger values leading to greater variability. The \theta, often referred to as the drift, introduces skewness; positive values of \theta skew the distribution to the right, while negative values skew it to the left. The r governs the tail heaviness and excess kurtosis; as r increases, the distribution more closely approximates a normal distribution with lighter tails and lower kurtosis. An alternative parameterization employs shape parameters \lambda > 0, \alpha > 0, and \beta \in \mathbb{R}, along with the location \mu \in \mathbb{R}, where the scale is implicitly defined via \gamma = \sqrt{\alpha^2 - \beta^2} to ensure the characteristic function is well-defined, subject to the constraint \alpha > |\beta|. In this form, \lambda relates to the kurtosis control, \beta drives the skewness, and \alpha influences the scale alongside \gamma. This parameterization is equivalent to the standard one, with relations such as r = 1/\nu = \lambda, \theta = \beta / \lambda, and \sigma = \sqrt{\alpha^2 - \beta^2} / \lambda. Note that some literature uses \nu = 1/r > 0 as the shape parameter, where small \nu (large r) approximates the normal distribution. The distribution has finite moments of all orders for all valid parameters. Positive \theta produces positive skewness, emphasizing right-tail outcomes, which is useful in modeling asset returns with occasional large gains. The distribution is supported on all real numbers x \in \mathbb{R}, and the parameters \sigma > 0 and r > 0 ensure the positive definiteness of the characteristic function, guaranteeing a valid probability distribution.

Probability Density Function

The probability density function of the variance-gamma distribution with location parameter \mu \in \mathbb{R}, scale parameter \sigma > 0, drift parameter \theta \in \mathbb{R}, and shape parameter r > 0 takes the closed-form expression p(x) = \frac{1}{\sigma \sqrt{\pi} \Gamma(r)} \exp\left( \frac{\theta (x - \mu)}{\sigma^2} \right) \left( \frac{|x - \mu|}{2 \sqrt{\theta^2 + \sigma^2}} \right)^{(r-1)/2} K_{(r-1)/2} \left( \frac{\sqrt{\theta^2 + \sigma^2}}{\sigma^2} |x - \mu| \right), where K_\nu(\cdot) denotes the modified Bessel function of the second kind of order \nu. This parameterization arises in financial modeling contexts, where \sigma controls the scale of Brownian motion components, \theta introduces asymmetry via drift, and r governs the kurtosis through the gamma time-change variance. An equivalent representation expresses the density as a normal variance-mean mixture, integrating over a gamma-distributed mixing variable g \sim \Gamma(r, 1/r): p(x; \mu, \sigma, \theta, r) = \int_0^\infty \frac{1}{\sqrt{2\pi g \sigma^2}} \exp\left( -\frac{(x - \mu - \theta g)^2}{2 g \sigma^2} \right) \frac{r^r g^{r-1} e^{-r g}}{\Gamma(r)} \, dg. This integral form highlights the distribution's construction as a Brownian motion with drift \theta and volatility \sigma, subordinated by a gamma process with mean rate 1 and variance rate $1/r. Evaluating the density requires computing the modified Bessel function K_{\nu}(z), which for small z uses the series expansion K_{\nu}(z) \approx \frac{1}{2} \Gamma(\nu) (z/2)^{-\nu} (valid as z \to 0^+) and for large z employs the asymptotic approximation K_{\nu}(z) \sim \sqrt{\pi / (2z)} e^{-z} (valid as z \to \infty). Software implementations facilitate this: the R package VarianceGamma provides the dVG function for direct density evaluation (using \nu = 1/r); in Python, the density can be computed using scipy.special.kv for the Bessel term combined with for the remaining components; MATLAB's Statistics and Machine Learning Toolbox includes besselk for the core computation, with user-defined wrappers for the full PDF. The integrates to 1 over \mathbb{R} by construction, as the representation convolves a properly normalized gamma with conditional densities that each integrate to 1. This property follows from the integral properties of the modified in the closed form, where \int_{-\infty}^\infty p(x) \, dx = 1 holds due to the derived from \Gamma functions and .

Moments and

Moments

The variance-gamma distribution VG(r, θ, σ, μ), parameterized by shape r > 0, drift θ ∈ ℝ, scale σ > 0, and location μ ∈ ℝ, has all moments finite due to its exponentially decaying tails, which ensure the integrability of |x|^k f(x) for any k > 0, where f is the . Higher-order moments can be obtained via differentiation of the or expressed in closed form using ratios of gamma functions and hypergeometric functions. The mean is given by \mathbb{E}[X] = \mu + r \theta. This follows from the expectation of the subordinated Brownian motion component, where the gamma subordinator has mean r. The variance is \text{Var}(X) = r (\sigma^2 + \theta^2). This expression arises from the applied to the representation X = μ + θ G + σ √G Z, where G ∼ Gamma(r, 1) (shape r, scale 1) has mean r and variance r, and Z ∼ N(0,1) independent of G. The (third standardized ) is \gamma_1 = \frac{\theta (3\sigma^2 + 2\theta^2)}{\sqrt{r} (\sigma^2 + \theta^2)^{3/2}}. In the symmetric case θ = 0, skewness vanishes. For θ > 0, the distribution is positively skewed, and vice versa. This is obtained by computing the third via successive differentiation of the and standardizing by the variance. The (fourth standardized ) is \beta_2 = 3 + \frac{3(\sigma^4 + 6\theta^2 \sigma^2 + 3\theta^4)}{r (\sigma^2 + \theta^2)^2}, yielding an excess kurtosis of \gamma_2 = \frac{3(\sigma^4 + 6\theta^2 \sigma^2 + 3\theta^4)}{r (\sigma^2 + \theta^2)^2}. In the symmetric case θ = 0, excess kurtosis simplifies to 3/r, which is 3 for r = 1 (corresponding to the Laplace distribution) and approaches 0 as r → ∞ (normal limit). These expressions are derived similarly from the fourth cumulant.

Characteristic Function

The characteristic function of a X following the variance-gamma distribution VG(r, θ, σ, μ) is \phi_X(t) = \exp\left(i t \mu \right) \left(1 - i \theta t + \frac{\sigma^2 t^2}{2}\right)^{-r}. This facilitates analytical tractability in probabilistic computations and was established in the foundational development of the model. In the alternative parametrization VG(σ, θ, ν) with ν = 1/r (where σ > 0 is the parameter, θ ∈ ℝ is the drift parameter, and ν > 0 is the variance rate of the subordinating ), it is given by \phi_X(t) = \left(1 - i \theta \nu t + \frac{\sigma^2 \nu}{2} t^2 \right)^{-1/\nu}. The moment-generating function follows analogously by replacing i t with z \in \mathbb{C}, M_X(z) = \exp\left(z \mu \right) \left(1 - \theta z + \frac{\sigma^2 z^2}{2}\right)^{-r}, defined and analytic in a strip of the complex plane containing a neighborhood of the origin, provided the parameters ensure the argument of the logarithm remains in the principal branch. The variance-gamma distribution emerges from the subordination of a with drift θ and diffusion coefficient σ by an independent with unit mean rate and variance 1/r, yielding the via the composition of their transforms. This construction underscores the model's role as a pure-jump when viewed in process form. For the associated , the characteristic exponent (or Lévy symbol) is \psi(t) = i t \mu - r \log\left(1 - i \theta t + \frac{\sigma^2 t^2}{2} \right), such that \phi_X(t) = e^{\psi(t)} for the unit-time increment; this form verifies the infinite divisibility of the distribution, as the exponent is well-defined for all t \in \mathbb{R}. In the ν parametrization, \psi(t) = -\frac{1}{\nu} \log\left(1 - i \theta \nu t + \frac{\sigma^2 \nu}{2} t^2 \right).

Limiting Cases

The variance-gamma (VG) distribution exhibits several limiting cases as its parameters approach specific boundary values, resulting in simpler well-known distributions. These limits highlight the VG distribution's role as a generalization of the while incorporating heavier tails and through its parameters. In the parameterization VG(μ, σ, θ, ν), where μ is the location, σ the scale, θ the asymmetry, and ν the parameter controlling the variance of the underlying gamma mixing distribution, the VG distribution converges to a as ν → 0. Specifically, it approaches N(μ + θ, σ² + θ² ν), which for small ν approximates N(μ + θ, σ²), reflecting the degeneration of the gamma time change to a Dirac delta at unity and recovering increments with drift θ and volatility σ. This limit underscores the VG's connection to Lévy processes subordinated by gamma time changes. When θ = 0, the VG distribution reduces to the symmetric variance-gamma distribution, centered at μ with equal tails. This symmetric case further specializes to the under particular parameter choices, such as ν = 1/(2 α²) in the scale-shape parameterization, yielding a double-exponential density that captures bilateral . The symmetric VG maintains the four-parameter structure but eliminates skewness, making it suitable for modeling symmetric heavy-tailed phenomena. In the alternative parameterization VG(μ, α, β, λ) involving the modified of kind, where α and β control the tail decay and skewness, and λ relates to the shape (often λ = 1/ν), the distribution approaches N(μ, 1/(2(α² - β²))) as λ → ∞. This convergence arises from the 's asymptotic behavior for large orders, smoothing the density toward Gaussianity. A notable simplification occurs when λ = 1/2, corresponding to ν = 2 in the previous parameterization; here, the reduces to an form due to the closed-form expression of the modified K_{1/2}(z) = \sqrt{\pi / (2z)} , e^{-z}. This yields the as a special case, with PDF proportional to \exp(-\alpha |x - \mu| + \beta (x - \mu)) adjusted by scale factors.

Connections to Other Distributions

The variance-gamma (VG) distribution belongs to the broader class of generalized hyperbolic (GH) distributions, specifically arising as a limiting case when the mixing parameter χ = 0 and the shape parameter λ > 0 in the GH parameterization. This embedding allows the VG to inherit the flexibility of the GH family in capturing asymmetry and heavy tails while simplifying to a four-parameter form suitable for applications requiring infinite activity jumps. In certain parameterizations of the GH family, the VG corresponds to the case where κ = -λ, reflecting the specific tail and kurtosis behavior distinct from other GH subclasses like the normal inverse Gaussian or hyperbolic distributions. A symmetric VG distribution with shape parameter λ = 1 and skewness parameter β = 0 reduces to the bilateral exponential distribution, also known as the . This connection highlights the VG's ability to generalize lighter-tailed distributions like the Laplace, which emerges as a special case exhibiting in both tails. The VG extends this by introducing additional parameters that allow for variable and asymmetry, making it a more versatile alternative in modeling phenomena with heavier tails than the Laplace but still finite variance under appropriate conditions. The VG distribution relates to the through the shared framework, where the skewed Student's t arises as a GH member with λ = -ν/2 and positive mixing parameter δ > 0, and the VG appears in the limit as δ → 0 while adjusting λ accordingly. This limiting or perspective positions the VG as a bridge between the polynomial-tailed Student's t and distributions with more pronounced jumps, often viewed as a variance with a gamma-distributed mixing variable that approximates t-like behavior for certain parameter regimes. As a , the VG process can be represented as a with drift subordinated by a , providing a time-change mechanism that generates the characteristic infinite activity and finite variation paths. This subordinator structure underscores its role in stochastic modeling, distinguishing it from pure processes while maintaining the property essential for Lévy-based financial models. Multivariate extensions of the VG distribution often involve mixing multivariate normals with gamma or generalized inverse Gaussian subordinators, and in matrix-variate forms, they can be constructed via Wishart-distributed mixing on the precision matrix to yield matrix VG distributions that preserve marginal VG properties and capture joint tail dependencies.

Properties and Behaviors

Convolution Properties

The class of variance-gamma (VG) distributions is closed under convolution, provided the independent random variables share the same skewness parameter \theta and scale parameter \sigma. If X_1 \sim \VG(r_1, \theta, \sigma, \mu_1) and X_2 \sim \VG(r_2, \theta, \sigma, \mu_2), then X_1 + X_2 \sim \VG(r_1 + r_2, \theta, \sigma, \mu_1 + \mu_2). In the alternative parametrization using \alpha, \beta, and \lambda, where the VG distribution arises as a normal variance-mean mixture with gamma mixing density of shape \lambda and rate $2/\nu, the sum has updated location \mu_1 + \mu_2 and \lambda = \lambda_1 + \lambda_2 while retaining the same \alpha and \beta. This closure extends to the reproduction property for sums of independent and identically distributed VG random variables. For n i.i.d. X_i \sim \VG(r, \theta, \sigma, \mu), the sum \sum_{i=1}^n X_i \sim \VG(n r, \theta, \sigma, n \mu). The property follows from the multiplicative structure of the under convolution, as detailed in the characteristic function section. The VG distribution is infinitely divisible, serving as the marginal distribution of a pure-jump with finite variation but infinite activity. Its infinite divisibility is characterized by the Lévy-Khinchin representation of the , with an explicit Lévy measure \nu(dx) = \frac{r}{|x|} \left( e^{-\lambda_+ |x|} \mathbf{1}_{x < 0}(x) + e^{-\lambda_- x} \mathbf{1}_{x > 0}(x) \right) dx, where \lambda_\pm = \frac{\sqrt{\theta^2 + \sigma^2} \pm \theta}{\sigma^2}. These properties enable the VG distribution to model aggregate risks and processes effectively, such as multi-period log-returns in pricing where daily increments are convolved to form longer-horizon .

Tail Behavior

The tail behavior of the variance-gamma (VG) exhibits a power-law modified , resulting in tails that are heavier than those of the normal distribution due to the infinite activity of the underlying but lighter than the power-law tails of α-stable distributions with index α < 2. This semi-heavy tail structure is essential for applications requiring realistic extreme value modeling, such as risk assessment in finance. For a VG random variable X with parameters λ (shape of the mixing gamma), α and β (related to the positive and negative jump intensities in the Lévy measure), the tail probability satisfies P(X > x) \sim c \, x^{\lambda - 1} e^{-\gamma x}, \quad x \to \infty, where c > 0 is a parameter-dependent constant and \gamma = \sqrt{\alpha^2 - \beta^2} governs the rate. This asymptotic form reflects the bilateral gamma construction of the VG distribution, where the tails decay exponentially but are tempered by a prefactor, distinguishing it from purely exponential or Gaussian tails. The heaviness of the tails is further evidenced by the excess kurtosis, which exceeds 3 for finite shape parameter λ (or equivalently, finite ν = 1/λ in alternative parametrizations), yielding a kurtosis greater than that of the normal distribution and indicating leptokurtic features. In the symmetric case (β = 0), the excess kurtosis simplifies to $3/\lambda, which increases as λ decreases, enhancing tail thickness. High quantiles of the VG distribution are driven by the decay of the modified of the second kind in the , whose large-argument asymptotic K_{\nu}(z) \sim \sqrt{\pi/(2z)} \, e^{-z} (for fixed order ν and z → ∞) directly contributes to the tail probability form above. If β > 0, the right tail becomes heavier relative to the left tail, as the decay rate γ decreases for positive deviations, making the VG distribution particularly suitable for capturing asymmetric positive jumps in processes like asset returns.

Applications

Financial Modeling

The variance-gamma (VG) distribution has been widely adopted in to represent asset returns, particularly for capturing the heavy tails, , and excess observed in empirical data that the Black-Scholes model fails to accommodate. In , the VG process models the log-price of an underlying asset as a with drift subordinated by a , enabling the derivation of European option prices in closed form through Fourier inversion of the . This approach leverages the analytically tractable of the VG process, φ(u) = (1 - iθνu + (σ²ν/2)u²)^{-t/ν}, to compute prices efficiently via techniques. Calibration of the VG model to financial data typically involves (MLE) applied to historical log-returns, optimizing parameters σ (), ν (variance of the gamma time change), and θ (drift) to fit the empirical . This decomposes the likelihood based on the VG , often using numerical optimization algorithms like BFGS or Nelder-Mead, and has shown superior performance over fits, especially for higher-frequency data such as daily or intraday returns. Empirical applications to stock returns from 2010–2020 demonstrate that VG parameters effectively capture time-varying and , with better log-likelihood and values in most cases. Compared to the Black-Scholes model, the VG process offers significant advantages by naturally generating volatility smiles and negative skewness in implied volatilities, aligning with observations in equity indices like the . Analysis of [S&P 500](/page/S&P 500) options from 1992–1994 reveals that VG corrects the strike and maturity biases inherent in Black-Scholes, with risk-neutral densities exhibiting negative skewness (θ ≈ -0.14) and kurtosis (ν ≈ 0.17), rejected in over 90% of Black-Scholes fits at the 1% significance level. In practice, the VG model supports and hedging in variance swaps by incorporating correlated VG processes to replicate variance exposures, as in the variance-gamma correlated (VGC) for optimal construction. For , extensions of the Merton structural model replace the normal return assumption with VG-distributed asset returns, yielding closed-form expressions for value-at-risk, , and entropic risk measures in portfolios involving mortgages and options. Empirical studies post-2008 crisis highlight the VG model's fit to high-frequency financial data. For instance, applications to intraday index data compare VG with alternatives like the normal-inverse Gaussian, assessing goodness-of-fit metrics for leptokurtic distributions observed during volatile periods. As of 2025, the VG model is implemented in open-source libraries like QuantLib, providing experimental engines for European option pricing under the VG process since version 1.1.

Other Fields

In , the wrapped variance-gamma distribution has been employed to model data, effectively capturing the circular statistics and intermittent fluctuations associated with turbulent flows, outperforming traditional von Mises distributions in fitting empirical observations from weather stations. The variance-gamma process has found application in for modeling non-monotonic degradation paths in mechanical components, such as the leakage rate in centrifugal pumps, allowing for accurate prognosis of remaining useful life and optimal imperfect maintenance scheduling under wear conditions. In recent advancements within (post-2020), the variance-gamma distribution emerges in analyses of generative models, where recursive training on causes parameter distributions to converge toward a variance-gamma form, highlighting mechanisms of model collapse and degradation in data quality over generations.

References

  1. [1]
  2. [2]
    The Variance Gamma (V.G.) Model for Share Market Returns - jstor
    Madan. University of Maryland. Eugene Seneta. University of Sydney. The Variance Gamma (V.G.). Model for Share Market. Returns*. I. Introduction. The purpose of ...
  3. [3]
    [PDF] The Variance Gamma Process and Option Pricing
    The process is obtained by evaluating Brownian motion with drift at a random time given by a gamma process.
  4. [4]
    The Variance Gamma (Vg) Model for Share Market Returns
    Aug 7, 2025 · Here, we use the following density function (Madan and Seneta, 1990;Seneta, 2004): ... ... variance gamma distribution for change point detection.
  5. [5]
    [PDF] The Variance-Gamma Distribution: A Review - arXiv
    Mar 9, 2023 · Madan and Seneta [77]. A number of basic properties of the ... [43] Gaunt, R. E. On the moments of the variance-gamma distribution.
  6. [6]
    [PDF] Wrapped variance gamma distribution with an application to Wind ...
    Aug 12, 2014 · Since some important variables are axial in weather study such as turbulent wind direction, the study of variance gamma distribution in case of ...
  7. [7]
    [PDF] Basics of Lévy processes∗ - Duke Economics
    Jun 9, 2012 · The idea of time deformation or subordination is due to Bochner (1949) and Bochner (1955), while it was introduced into economics by Clark ...
  8. [8]
    [PDF] lévy processes, stable processes, and subordinators
    The fundamental Lévy processes are the Wiener process and the Poisson process. The Poisson process is a subordinator, but is not stable; the Wiener process is.
  9. [9]
    Subordinated Levy Processes and Applications to Crude Oil Options in
    Sep 1, 2005 · Clark (1973) proposed Bochner's (1955) concept of subordinated stochastic process as a model to account for non-normality of returns. He showed ...
  10. [10]
    Variance Gamma Process and Option Pricing | Review of Finance
    Published: 01 April 1998. PDF. Views. Article contents. Cite. Cite. Dilip B. Madan, Peter P. Carr, Eric C. Chang, The Variance Gamma Process and Option ...
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
    [2303.05615] The Variance-Gamma Distribution: A Review - arXiv
    Mar 9, 2023 · The variance-gamma (VG) distributions form a four-parameter family which includes as special and limiting cases the normal, gamma and Laplace distributions.
  16. [16]
    [PDF] A Class of Non-Gaussian Second Order Random Fields
    Due to this representation the Laplace motion is often called variance-gamma process, since ... In the limiting case when ν → 0 one will get the Gaussian ...
  17. [17]
    [1309.4422] Variance-Gamma approximation via Stein's method
    Sep 17, 2013 · Variance-Gamma distributions are widely used in financial modelling and contain as special cases the normal, Gamma and Laplace distributions. In ...
  18. [18]
    [PDF] ghyp: A package on generalized hyperbolic distributions - CRAN
    If χ = 0 and λ > 0 one gets a limiting case which is known amongst others as Variance. Gamma (VG) distribution. • If ψ = 0 and λ < 0 the generalized hyperbolic ...
  19. [19]
    The Variance-Gamma Distribution: A Review - Project Euclid
    The variance-gamma (VG) distributions form a four-parameter family which includes as special and limiting cases the normal, gamma and Laplace distributions.Missing: original | Show results with:original
  20. [20]
    On the relationship between the generalized student t-distribution ...
    Sep 5, 2012 · On the relationship between the generalized student t-distribution and the variance gamma distribution in statistical analysis of random-size ...
  21. [21]
    (PDF) Robust Kalman Filters Based on the Sub-Gaussianα-stable ...
    ... κ(λk))}and ε2is a small positive. number. In addition, the fixed-point ... variance Gamma. distribution (RKF-VG) [16] and the RSTKF [13]. Lastly, our.
  22. [22]
    An ECM algorithm for Skewed Multivariate Variance ... - NASA ADS
    Especially for distributions with normal mean-variance mixtures representation such as the multivariate skewed variance gamma (MSVG) distribution, it utilises ...<|control11|><|separator|>
  23. [23]
    [PDF] The Variance Gamma Process and Option Pricing
    Madan, Dilip B. and Eugene Seneta (1990) The variance gamma (V.G.) model for share market returns, Journal of Business 63(4), 511– ...
  24. [24]
    [PDF] Variance Gamma process in the option pricing model
    The variable 𝜔 can be derived from a characteristic function of Variance Gamma distribution ... We can obtain characteristic function for the Variance Gamma model.
  25. [25]
    An Empirical Examination of the Variance-Gamma Model for Foreign ...
    Madan, D., and E. Seneta. 1990. The variance gamma model for share market returns. Journal of Business 63:511–24. Merton, R. C. 1976. Option pricing when ...
  26. [26]
    [PDF] Optimal Variance Swaps Investments - IAENG
    Abstract—In this paper, we focus on constructing optimal portfolios of variance swaps based on a vari- ance Gamma correlated (VGC) model. Each variance.<|separator|>
  27. [27]
    A Credit-Risk Valuation under the Variance-Gamma Asset Return
    We suggest that the returns on stock are variance-gamma distributed. The value at risk, the expected shortfall and the entropic risk measure for this portfolio ...
  28. [28]
    [PDF] bradscholars - Bradford Scholars
    In the following subsections, we introduce the variance-gamma process and extend the Merton structural credit risk model via the variance-gamma convolution.
  29. [29]
    Variance-Gamma and Normal-Inverse Gaussian models: Goodness
    For high-frequency Chinese index returns, NIG model is more robust and provides a better fit to the empirical distributions of returns at different time scales.
  30. [30]
    QuantLib: Version history - kapl.org redirected
    Variance-gamma implementation (thanks to Adrian O'Neill.) Contributed classes include a variance-gamma process and model (with data but no behavior at this ...<|control11|><|separator|>
  31. [31]
    Variance Gamma process as degradation model for prognosis and ...
    Variance Gamma Process (VGP) as a non-monotonous degradation model. · Failure time estimation and sensitivity analysis. · Prognosis and uncertainty analysis.
  32. [32]
    [1510.02302] Dependent time changed processes with applications ...
    Oct 8, 2015 · Dependent time changed processes with applications to nonlinear ocean waves. Authors:Pierre Ailliot, Bernard Delyon, Valérie Monbet, Marc ...