Yule–Simon distribution

Parameters Probability mass function Yule–Simon PMF on a log-log scale. (Note that the function is only defined at integer values of k. The connecting lines do not indicate continuity.) Cumulative distribution function Yule–Simon CMF. (Note that the function is only defined at integer values of k. The connecting lines do not indicate continuity.) $\rho >0\,$ shape (real) $k\in \{1,2,\dotsc \}$ $\rho \operatorname {B} (k,\rho +1)$ $1-k\operatorname {B} (k,\rho +1)$ ${\frac {\rho }{\rho -1}}$ for $\rho >1$ $1$ ${\frac {\rho ^{2}}{(\rho -1)^{2}(\rho -2)}}$ for $\rho >2$ ${\frac {(\rho +1)^{2}{\sqrt {\rho -2}}}{(\rho -3)\rho }}\,$ for $\rho >3$ $\rho +3+{\frac {11\rho ^{3}-49\rho -22}{(\rho -4)(\rho -3)\rho }}$ for $\rho >4$ ${\frac {\rho }{\rho +1}}{}_{2}F_{1}(1,1;\rho +2;e^{t})\,e^{t}$ ${\frac {\rho }{\rho +1}}{}_{2}F_{1}(1,1;\rho +2;e^{i\,t})e^{i\,t}$ In probability and statistics, the Yule–Simon distribution is a discrete probability distribution named after Udny Yule and Herbert A. Simon. Simon originally called it the Yule distribution.

The probability mass function (pmf) of the Yule–Simon (ρ) distribution is

$f(k;\rho )=\rho \operatorname {B} (k,\rho +1),$ for integer $k\geq 1$ and real $\rho >0$ , where $\operatorname {B}$ is the beta function. Equivalently the pmf can be written in terms of the falling factorial as

$f(k;\rho )={\frac {\rho \Gamma (\rho +1)}{(k+\rho )^{\underline {\rho +1}}}},$ where $\Gamma$ is the gamma function. Thus, if $\rho$ is an integer,

$f(k;\rho )={\frac {\rho \,\rho !\,(k-1)!}{(k+\rho )!}}.$ The parameter $\rho$ can be estimated using a fixed point algorithm.

The probability mass function f has the property that for sufficiently large k we have

$f(k;\rho )\approx {\frac {\rho \Gamma (\rho +1)}{k^{\rho +1}}}\propto {\frac {1}{k^{\rho +1}}}.$ This means that the tail of the Yule–Simon distribution is a realization of Zipf's law: $f(k;\rho )$ can be used to model, for example, the relative frequency of the $k$ th most frequent word in a large collection of text, which according to Zipf's law is inversely proportional to a (typically small) power of $k$ .

Occurrence

The Yule–Simon distribution arose originally as the limiting distribution of a particular stochastic process studied by Yule as a model for the distribution of biological taxa and subtaxa. Simon dubbed this process the "Yule process" but it is more commonly known today as a preferential attachment process.[citation needed] The preferential attachment process is an urn process in which balls are added to a growing number of urns, each ball being allocated to an urn with probability linear in the number the urn already contains.

The distribution also arises as a compound distribution, in which the parameter of a geometric distribution is treated as a function of random variable having an exponential distribution.[citation needed] Specifically, assume that $W$ follows an exponential distribution with scale $1/\rho$ or rate $\rho$ :

$W\sim \operatorname {Exponential} (\rho ),$ with density

$h(w;\rho )=\rho \exp(-\rho w).$ Then a Yule–Simon distributed variable K has the following geometric distribution conditional on W:

$K\sim \operatorname {Geometric} (1-\exp(-W))\,.$ The pmf of a geometric distribution is

$g(k;p)=p(1-p)^{k-1}$ for $k\in \{1,2,\dotsc \}$ . The Yule–Simon pmf is then the following exponential-geometric compound distribution:

$f(k;\rho )=\int _{0}^{\infty }g(k;\exp(-w))h(w;\rho )\,dw.$ The maximum likelihood estimator for the parameter $\rho$ given the observations $k_{1},k_{2},k_{3},\dots ,k_{N}$ is the solution to the fixed point equation

$\rho ^{(t+1)}={\frac {N+a-1}{b+\sum _{i=1}^{N}\sum _{j=1}^{k_{i}}{\frac {1}{\rho ^{(t)}+j}}}},$ where $b=0,a=1$ are the rate and shape parameters of the gamma distribution prior on $\rho$ .

This algorithm is derived by Garcia  by directly optimizing the likelihood. Roberts and Roberts 

generalize the algorithm to Bayesian settings with the compound geometric formulation described above. Additionally, Roberts and Roberts  are able to use the Expectation Maximisation (EM) framework to show convergence of the fixed point algorithm. Moreover, Roberts and Roberts  derive the sub-linearity of the convergence rate for the fixed point algorithm. Additionally, they use the EM formulation to give 2 alternate derivations of the standard error of the estimator from the fixed point equation. The variance of the $\lambda$ estimator is

$\operatorname {Var} ({\hat {\lambda }})={\frac {1}{{\frac {N}{{\hat {\lambda }}^{2}}}-\sum _{i=1}^{N}\sum _{j=1}^{k_{i}}{\frac {1}{({\hat {\lambda }}+j)^{2}}}}},$ the standard error is the square root of the quantity of this estimate divided by N.

Generalizations

The two-parameter generalization of the original Yule distribution replaces the beta function with an incomplete beta function. The probability mass function of the generalized Yule–Simon(ρ, α) distribution is defined as

$f(k;\rho ,\alpha )={\frac {\rho }{1-\alpha ^{\rho }}}\;\mathrm {B} _{1-\alpha }(k,\rho +1),\,$ with $0\leq \alpha <1$ . For $\alpha =0$ the ordinary Yule–Simon(ρ) distribution is obtained as a special case. The use of the incomplete beta function has the effect of introducing an exponential cutoff in the upper tail.

Computer software for the Yule-Simon distribution

Yule-Simon distribution using Python

The Yule-Simon distribution is available in the python scipy library. 

Bibliography

• Colin Rose and Murray D. Smith, Mathematical Statistics with Mathematica. New York: Springer, 2002, ISBN 0-387-95234-9. (See page 107, where it is called the "Yule distribution".)