Stable distribution
≈ 14 min read
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
In probability theory, a distribution is said to be stable if a linear combination of two independent random variables with this distribution has the same distribution, up to location and scale parameters. A random variable is said to be stable if its distribution is stable. The stable distribution family is also sometimes referred to as the Lévy alpha-stable distribution, after Paul Lévy, the first mathematician to have studied it. Of the four parameters defining the family, most attention has been focused on the stability parameter, \alpha (see panel).
Stable distributions have 0 , with the upper bound corresponding to the normal distribution, and approaches the Dirac delta function in the limit as \alpha \rightarrow 0 . The distributions have undefined variance for \alpha , and undefined mean for \alpha \leq 1 . The importance of stable probability distributions is that they are "attractors" for properly normed sums of independent and identically distributed (iid) random variables. The normal distribution defines a family of stable distributions.
By the classical central limit theorem, the properly normed sum of a set of random variables, each with finite variance, will tend toward a normal distribution as the number of variables increases. Without the finite variance assumption, the limit may be a stable distribution that is not normal.…
In probability theory, a distribution is said to be stable if a linear combination of two independent random variables with this distribution has the same distribution, up to location and scale parameters. A random variable is said to be stable if its distribution is stable. The stable distribution family is also sometimes referred to as the Lévy alpha-stable distribution, after Paul Lévy, the first mathematician to have studied it. Of the four parameters defining the family, most attention has been focused on the stability parameter, \alpha (see panel).
Stable distributions have 0 , with the upper bound corresponding to the normal distribution, and approaches the Dirac delta function in the limit as \alpha \rightarrow 0 . The distributions have undefined variance for \alpha , and undefined mean for \alpha \leq 1 . The importance of stable probability distributions is that they are "attractors" for properly normed sums of independent and identically distributed (iid) random variables. The normal distribution defines a family of stable distributions.
By the classical central limit theorem, the properly normed sum of a set of random variables, each with finite variance, will tend toward a normal distribution as the number of variables increases. Without the finite variance assumption, the limit may be a stable distribution that is not normal. Mandelbrot referred to such distributions as "stable Paretian distributions", after Vilfredo Pareto.
In particular, he referred to those maximally skewed in the positive direction with 1 as "Pareto–Lévy distributions", which he regarded as better descriptions of stock and commodity prices than normal distributions.
Definition
A non-degenerate distribution is a stable distribution if it satisfies the following property: Since the normal distribution, the Cauchy distribution, and the Lévy distribution all have the above property, it follows that they are special cases of stable distributions.
Such distributions form a four-parameter family of continuous probability distributions parametrized by location and scale parameters μ and c, respectively, and two shape parameters \beta and \alpha , roughly corresponding to measures of asymmetry and concentration, respectively (see the figures). The characteristic function \varphi of a probability distribution with density function f is the Fourier transform of f.
The density function is then the inverse Fourier transform of the characteristic function: \varphi(t) = \int_{- \infty}^\infty f(x)e^{ ixt}\,dx. Although the probability density function for a general stable distribution cannot be written analytically, the general characteristic function can be expressed analytically.
A random variable X is called stable if its characteristic function can be written as \varphi(t; \alpha, \beta, c, \mu) = \exp \left ( i t \mu - |c t|^\alpha \left ( 1 - i \beta \sgn(t) \Phi \right ) \right ) where is just the sign of and \Phi = \begin{cases} \tan \left (\frac{\pi \alpha}{2} \right) & \alpha \neq 1 \\ - \frac{2}{\pi}\log|t| & \alpha = 1 \end{cases} μ ∈ R is a shift parameter, \beta \in [-1,1] , called the skewness parameter, is a measure of asymmetry.
Notice that in this context the usual skewness is not well defined, as for \alpha the distribution does not admit 2nd or higher moments, and the usual skewness definition is the 3rd central moment. The reason this gives a stable distribution is that the characteristic function for the sum of two independent random variables equals the product of the two corresponding characteristic functions. Adding two random variables from a stable distribution gives something with the same values of \alpha and \beta , but possibly different values of μ and c.
Not every function is the characteristic function of a legitimate probability distribution (that is, one whose cumulative distribution function is real and goes from 0 to 1 without decreasing), but the characteristic functions given above will be legitimate so long as the parameters are in their ranges. The value of the characteristic function at some value t is the complex conjugate of its value at −t as it should be so that the probability distribution function will be real.
In the simplest case \beta = 0 , the characteristic function is just a stretched exponential function; the distribution is symmetric about μ and is referred to as a (Lévy) symmetric alpha-stable distribution, often abbreviated SαS. When \alpha and \beta = 1 , the distribution is supported on [μ, ∞). The parameter c > 0 is a scale factor which is a measure of the width of the distribution while \alpha is the exponent or index of the distribution and specifies the asymptotic behavior of the distribution.
The parametrization above is easiest to use for theoretical work, but its probability density is not continuous in the parameters at \alpha =1 .
A continuous parametrization, better for numerical work, is \varphi(t; \alpha, \beta, \gamma, \delta) = \exp \left (i t \delta - |\gamma t|^\alpha \left (1 - i \beta \sgn(t) \Phi \right ) \right ) where: \Phi = \begin{cases} \left ( 1 - |\gamma t|^{1 - \alpha} \right ) \tan \left (\tfrac{\pi \alpha}{2} \right ) & \alpha \neq 1 \\[1ex] - \frac{2}{\pi} \log|\gamma t| & \alpha = 1 \end{cases} The ranges of \alpha and \beta are the same as before, γ (like c) should be positive, and δ (like μ) should be real.
In either parametrization one can make a linear transformation of the random variable to get a random variable whose density is In the first parametrization ("1"), this is done by defining the new variable: y = \begin{cases} \frac{x - \mu}c & \alpha \neq 1 \\[1ex] \frac{x - \mu}c - \beta\frac 2\pi\ln c & \alpha = 1 \end{cases} For the second parametrization ("0"), simply use y = \frac{x-\delta}\gamma independent of \alpha .
In the first parametrization, if the mean exists (that is, \alpha > 1 ) then it is equal to μ, whereas in the second parametrization when the mean exists it is equal to \delta - \beta \gamma \tan \left (\tfrac{\pi\alpha}{2} \right).
The distribution
A stable distribution is therefore specified by the above four parameters. It can be shown that any non-degenerate stable distribution has a smooth (infinitely differentiable) density function.
If f(x; \alpha, \beta, c, \mu) denotes the density of X and Y is the sum of independent copies of X: Y = \sum_{i = 1}^N k_i (X_i - \mu) then Y has the density \tfrac{1}{s} f(y / s; \alpha, \beta, c, 0) with s = \left(\sum_{i = 1}^N |k_i|^\alpha \right )^{\frac{1}{\alpha}} The asymptotic behavior is described, for \alpha , by: f(x) \sim \frac{1} Some of the special cases are known by particular names: For \alpha = 1 and \beta = 1 , the distribution is a Landau distribution (L) which has a specific usage in physics under this name.
For \alpha = 3/2 and \beta = 0 the distribution reduces to a Holtsmark distribution with scale parameter c and shift parameter μ. Also, in the limit as c approaches zero or as α approaches zero the distribution will approach a Dirac delta function .
Series representation
The stable distribution can be restated as the real part of a simpler integral: f(x;\alpha,\beta,c,\mu)=\frac{1}{\pi}\Re\left[ \int_0^\infty e^{it(x-\mu)}e^{-(ct)^\alpha(1-i\beta\Phi)}\,dt\right]. Expressing the second exponential as a Taylor series, this leads to: f(x;\alpha,\beta,c,\mu)=\frac{1}{\pi}\Re\left[ \int_0^\infty e^{it(x-\mu)}\sum_{n=0}^\infty\frac{(-qt^\alpha)^n}{n!}\,dt\right] where q=c^\alpha(1-i\beta\Phi) .
Reversing the order of integration and summation, and carrying out the integration yields: f(x;\alpha,\beta,c,\mu)=\frac{1}{\pi}\Re\left[ \sum_{n=1}^\infty\frac{(-q)^n}{n!}\left(\frac{i}{x-\mu}\right)^{\alpha n+1}\Gamma(\alpha n+1)\right] which will be valid for x ≠ μ and will converge for appropriate values of the parameters. (Note that the n = 0 term which yields a delta function in x − μ has therefore been dropped.) Expressing the first exponential as a series will yield another series in positive powers of x − μ which is generally less useful.
For one-sided stable distribution, the above series expansion needs to be modified, since q=\exp(-i\alpha\pi/2) and q i^{\alpha}=1 . There is no real part to sum.
Instead, the integral of the characteristic function should be carried out on the negative axis, which yields: \begin{align} L_\alpha(x) & = \frac{1}{\pi}\Re\left[ \sum_{n=1}^\infty\frac{(-q)^n}{n!}\left(\frac{-i}{x}\right)^{\alpha n+1}\Gamma(\alpha n+1)\right] \\[1ex] & = \frac{1}{\pi}\sum_{n=1}^\infty\frac{-\sin(n(\alpha+1)\pi)}{n!}\left(\frac{1}{x}\right)^{\alpha n+1}\Gamma(\alpha n+1) \end{align} for \alpha\ne 1 compute: X = \left (1+\zeta^2 \right )^\frac{1}{2\alpha} \frac{\sin ( \alpha(U+\xi)) }{ (\cos(U))^{\frac{1}{\alpha}}} \left (\frac{\cos (U - \alpha(U+\xi)) }{W} \right )^\frac{1-\alpha}{\alpha}, for \alpha=1 compute: X = \frac{1}{\xi}\left\{\left(\frac{\pi}{2}+\beta U \right)\tan U- \beta\log\left(\frac{\frac{\pi}{2} W\cos U}{\frac{\pi}{2}+\beta U}\right)\right\}, where \zeta = -\beta\tan\frac{\pi\alpha}{2}, \qquad \xi =\begin{cases} \frac{1}{\alpha} \arctan(-\zeta) & \alpha \ne 1 \\ \frac{\pi}{2} & \alpha=1 \end{cases} This algorithm yields a random variable X\sim S_\alpha(\beta,1,0) .
For a detailed proof see. To simulate a stable random variable for all admissible values of the parameters \alpha , c , \beta and \mu use the following property: If X \sim S_\alpha(\beta,1,0) then Y = \begin{cases} c X+\mu & \alpha \ne 1 \\ c X+\mu + \frac{2}{\pi}\beta c\log c & \alpha = 1 \end{cases} is S_\alpha(\beta,c,\mu) . For \alpha = 2 (and \beta = 0 ) the CMS method reduces to the well known Box-Muller transform for generating Gaussian random variables.
While other approaches have been proposed in the literature, including application of Bergström and LePage series expansions, the CMS method is regarded as the fastest and the most accurate.
Let S_{\mu,\nu}(z) be a Lommel function, then: f {\left (x;\tfrac{1}{3},0,1,0\right )} = \Re\left ( \frac{2e^{- \frac{i \pi}{4}}}{3 \sqrt{3} \pi} \frac{1}{\sqrt{x^3}} S_{0,\frac{1}{3}} {\left (\frac{2e^{\frac{i \pi}{4}}}{3 \sqrt{3}} \frac{1}{\sqrt{x}} \right)} \right ) Let S(x) and C(x) denote the Fresnel integrals, then: f{\left (x;\tfrac{1}{2},0,1,0\right )} = \left(\tfrac{1}{2\pi\left|x\right|^3}\right)^{1/2} \left (\sin\left(\tfrac{1}{4|x|}\right) \left [\tfrac{1}{2} - S{\left (\tfrac{1}{\sqrt{2\pi|x|}}\right )}\right ] + \cos\left(\tfrac{1}{4|x|} \right) \left [\tfrac{1}{2} - C{\left (\tfrac{1}{\sqrt{2\pi|x|}}\right )}\right ]\right ) Let K_v(x) be the modified Bessel function of the second kind, then: f{\left (x;\tfrac{1}{3},1,1,0\right )} = \frac{2^{\frac{5}{2}}}{3^{\frac{7}{4}}\pi} \frac{1}{\sqrt{x^3}} K_{\frac{1}{3}}{\left (\frac{2^{\frac{5}{2}}}{3^{\frac{9}{4}}} \frac{1}{\sqrt{x}} \right )} Let {}_mF_n denote the hypergeometric functions, then: \begin{align} f{\left(x; \tfrac{4}{3}, 0, 1, 0\right)} &= \frac{3^{\frac{5}{4}}}{2^{\frac{5}{2}} \pi^{\frac{1}{2}}} \frac{\Gamma{\left (\tfrac{7}{12} \right )} \, \Gamma{\left(\tfrac{11}{12} \right )}}{\Gamma{\left( \tfrac{6}{12} \right)} \, \Gamma{\left (\tfrac{8}{12} \right )}} \; {}_2F_2{\left ( \tfrac{7}{12}, \tfrac{11}{12}; \tfrac{6}{12}, \tfrac{8}{12}; \tfrac{3^3 x^4}{4^4} \right)} \\[2pt] & \quad - \frac{3^{\frac{11}{4}} x^3}{2^{\frac{13}{2}} \pi^{\frac{1}{2}}} \frac{\Gamma{\left(\tfrac{13}{12} \right)} \, \Gamma{\left (\tfrac{17}{12} \right )}}{\Gamma{\left (\tfrac{18}{12} \right )} \, \Gamma{\left (\tfrac{15}{12} \right )}} \; {}_2F_2 {\left( \tfrac{13}{12}, \tfrac{17}{12}; \tfrac{18}{12}, \tfrac{15}{12}; \tfrac{3^3 x^4}{4^4} \right)} \end{align} \begin{align} f{\left(x; \tfrac{3}{2}, 0, 1, 0\right)} &= \frac{\Gamma{\left(\tfrac{5}{3} \right)}}{\pi} {}_2F_3 {\left( \tfrac{5}{12}, \tfrac{11}{12}; \tfrac{1}{3}, \tfrac{1}{2}, \tfrac{5}{6}; - \tfrac{2^2 x^6}{3^6} \right)} \\[2pt] & \quad - \frac{x^2}{3 \pi} \, {}_3F_4 {\left( \tfrac{3}{4}, 1, \tfrac{5}{4}; \tfrac{2}{3}, \tfrac{5}{6}, \tfrac{7}{6}, \tfrac{4}{3}; - \tfrac{2^2 x^6}{3^6} \right)} \\[2pt] & \quad + \frac{7 x^4 \Gamma{\left(\tfrac{4}{3} \right)}}{3^4 \pi ^ 2} {}_2F_3{\left ( \tfrac{13}{12}, \tfrac{19}{12}; \tfrac{7}{6}, \tfrac{3}{2}, \tfrac{5}{3}; -\tfrac{2^2 x^6}{3^6} \right)} \end{align} with the latter being the Holtsmark distribution.
Let W_{k,\mu}(z) be a Whittaker function, then: \begin{align} f\left (x;\tfrac{2}{3},0,1,0\right ) &= \frac{\sqrt{3}}{6\sqrt{\pi}|x|} \exp\left (\tfrac{2}{27}x^{-2}\right ) W_{-\frac{1}{2},\frac{1}{6}}\left (\tfrac{4}{27}x^{-2}\right ) \\[8pt] f\left (x;\tfrac{2}{3},1,1,0\right ) &= \frac{\sqrt{3}}{\sqrt{\pi}|x|} \exp\left (-\tfrac{16}{27}x^{-2}\right ) W_{\frac{1}{2},\frac{1}{6}} \left (\tfrac{32}{27}x^{-2}\right ) \\[8pt] f\left (x;\tfrac{3}{2},1,1,0\right ) &= \begin{cases} \frac{\sqrt{3}}{\sqrt{\pi}|x|} \exp\left (\frac{1}{27}x^3\right ) W_{\frac{1}{2},\frac{1}{6}}\left (- \frac{2}{27}x^3\right ) & x
See also
Lévy flight Lévy process Other "power law" distributions Pareto distribution Zeta distribution Zipf distribution Zipf–Mandelbrot distribution Financial models with long-tailed distributions and volatility clustering Multivariate stable distribution Discrete-stable distribution
References
Text derived from https://en.wikipedia.org/w/index.php?oldid=1377309917 under CC BY-SA 4.0; modifications have been made.