cumulant
set of quantities that provide an alternative to the moments of the distribution in probability theorie
≈ 25 min read
Names in other languages
- РусскийПолуинвариант
- DeutschKumulante
- 日本語キュムラント母関数
- PolskiKumulanta
- Portuguêscumulante
- slKumulanta
- Englishcumulant
- Françaiscumulant
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
In probability theory and statistics, the cumulants of a probability distribution are a set of quantities that provide an alternative to the moments of the distribution. Any two probability distributions whose moments are identical will have identical cumulants as well, and vice versa.
The first cumulant is the mean, the second cumulant is the variance, and the third cumulant is the same as the third central moment. But fourth and higher-order cumulants are not equal to central moments. In some cases theoretical treatments of problems in terms of cumulants are simpler than those using moments.
In particular, when two or more random variables are statistically independent, the th-order cumulant of their sum is equal to the sum of their th-order cumulants. Also, the third and higher-order cumulants of a normal distribution are zero, and it is the only distribution with this property.
Just as for moments, where joint moments are used for collections of random variables, it is possible to define joint cumulants. K(t)=\log\operatorname{E}\left[e^{tX}\right].
The cumulants are obtained from a power series expansion of the cumulant generating function: K(t)=\sum_{n=1}^\infty \kappa_{n} \frac{t^{n}}{n!} =\kappa_1 \frac{t}{1!} + \kappa_2 \frac{t^2}{2!}+ \kappa_3 \frac{t^3}{3!}+ \cdots = \mu t + \sigma^2 \frac{t^2}{2} + \cdots. This expansion is a Maclaurin series, so the th cumulant can be obtained by differentiating the above expansion times and evaluating the result at zero: \kappa_{n} = K^{(n)}(0).
If the moment-generating function does not exist, the cumulants can be defined in terms of the relationship between cumulants and moments discussed later. H(t)=\log\operatorname{E} \left[e^{i t X}\right]=\sum_{n=1}^\infty \kappa_n \frac{(it)^n}{n!}=\mu it - \sigma^2 \frac{ t^2}{2} + \cdots An advantage of — in some sense the function evaluated for purely imaginary arguments — is that is well defined for all real values of even when is not well defined for all real values of , such as can occur when there is "too much" probability that has a large magnitude.
Although the function will be well defined, it will nonetheless mimic in terms of the length of its Maclaurin series, which may not extend beyond (or, rarely, even to) linear order in the argument , and in particular the number of cumulants that are well defined will not change. Nevertheless, even when does not have a long Maclaurin series, it can be used directly in analyzing and, particularly, adding random variables.
Both the Cauchy distribution (also called the Lorentzian) and more generally, stable distributions (related to the Lévy distribution) are examples of distributions for which the power-series expansions of the generating functions have only finitely many well-defined terms. The n th cumulant \kappa_n(X) of (the distribution of) a random variable X enjoys the following properties: If n>1 and c is constant (i.e. not random) then \kappa_n(X+c) = \kappa_n(X), i.e. the cumulant is translation invariant.
(If n=1 then we have \kappa_1(X+c) = \kappa_1(X)+c.) If c is constant (i.e. not random) then \kappa_n(cX) = c^n\kappa_n(X), i.e. the n th cumulant is homogeneous of degree n . If random variables X_1,\ldots,X_m are independent then \kappa_n(X_1+\cdots+X_m) = \kappa_n(X_1) + \cdots + \kappa_n(X_m)\,. That is, the cumulant is cumulative — hence the name.
The cumulative property follows quickly by considering the cumulant generating function: \begin{align} K_{X_1+\cdots+X_m}(t) & =\log\operatorname{E} \left[e^{t(X_1+\cdots+X_m)}\right] \\[5pt] & = \log \left(\operatorname{E} \left[e^{tX_1}\right] \cdots \operatorname{E} \left[ e^{tX_m} \right] \right) \\[5pt] & = \log\operatorname{E}\left[e^{tX_1}\right] + \cdots + \log \operatorname{E} \left[ e^{tX_m} \right] \\[5pt] &= K_{X_1}(t) + \cdots + K_{X_m}(t), \end{align} so that each cumulant of a sum of independent random variables is the sum of the corresponding cumulants of the addends. That is, when the addends are statistically independent, the mean of the sum is the sum of the means, the variance of the sum is the sum of the variances, the third cumulant (which happens to be the third central moment) of the sum is the sum of the third cumulants, and so on for each order of cumulant.
A distribution with given cumulants can be approximated through an Edgeworth series. Let \kappa_n(X) be the cumulants, m(X) := \operatorname E\left[X\right] be the mean, and \mu_n(X) := \operatorname E\left[{\left(X-\operatorname E[X]\right)}^n\right] ={} be the central moments.
Then: \kappa_1 = m = \operatorname{mean}(X) (the mean). \kappa_2 = \mu_2 = \operatorname{var}(X) (the variance, or second central moment). \kappa_3 = \mu_3 . \kappa_4 = \mu_4 - 3\mu_2^2 . (This is the first case in which cumulants are not simply moments or central moments. The central moments of degree more than 3 lack the cumulative property.) \kappa_5 = \mu_5 - 10 \mu_3 \mu_2 .
The Bernoulli distributions, (number of successes in one trial with probability of success). The cumulant generating function is . The first cumulants are and .
The cumulants satisfy a recursion formula \kappa_{n+1}=p (1-p) \frac{d\kappa_n}{dp}. The geometric distributions, (number of failures before one success with probability of success on each trial). The cumulant generating function is .
The first cumulants are , and . Substituting gives and . The Poisson distributions.
The cumulant generating function is . All cumulants are equal to the parameter: . The binomial distributions, (number of successes in independent trials with probability of success on each trial).
The special case is a Bernoulli distribution. Every cumulant is just times the corresponding cumulant of the corresponding Bernoulli distribution. The cumulant generating function is .
The first cumulants are and . Substituting gives and . The limiting case is a Poisson distribution.
The negative binomial distributions, (number of failures before successes with probability of success on each trial). The special case is a geometric distribution. Every cumulant is just times the corresponding cumulant of the corresponding geometric distribution.
The derivative of the cumulant generating function is . The first cumulants are , and . Substituting gives and .
Comparing these formulas to those of the binomial distributions explains the name 'negative binomial distribution'. The limiting case is a Poisson distribution.
Introducing the variance-to-mean ratio \varepsilon=\mu^{-1}\sigma^2=\kappa_1^{-1}\kappa_2, the above probability distributions get a unified formula for the derivative of the cumulant generating function: K'(t)=(1+(e^{-t}-1)\varepsilon)^{-1}\mu The second derivative is K(t)=(\varepsilon-(\varepsilon-1)e^t)^{-2}\mu\varepsilon e^t confirming that the first cumulant is and the second cumulant is . The constant random variables have .
The binomial distributions have so that . The Poisson distributions have . The negative binomial distributions have so that .
Note the analogy to the classification of conic sections by eccentricity: circles , ellipses , parabolas , hyperbolas . \begin{align} & \exists c>0,\,\, F(x)=O(e^{cx}), x\to-\infty; \text{ and} \\[4pt] & \exists d>0,\,\, 1-F(x)=O(e^{-dx}),x\to+\infty; \end{align} where F is the cumulative distribution function. The cumulant generating function will have vertical asymptote(s) at the negative supremum of such , if such a supremum exists, and at the supremum of such , if such a supremum exists, otherwise it will be defined for all real numbers.
If the support of a random variable has finite upper or lower bounds, then its cumulant generating function , if it exists, approaches asymptote(s) whose slope is equal to the supremum or infimum of the support, \begin{align} y & =(t+1)\inf \operatorname{supp} X-\mu(X), \text{ and} \\[5pt] y & =(t-1)\sup \operatorname{supp}X+\mu(X), \end{align} respectively, lying above both these lines everywhere. (The integrals \int_{-\infty}^0 \left[t\inf \operatorname{supp}X-K'(t)\right]\,dt, \qquad \int_{\infty}^0 \left[t\inf \operatorname{supp}X-K'(t) \right]\,dt yield the -intercepts of these asymptotes, since .)
For a shift of the distribution by , K_{X+c}(t)=K_X(t)+ct. For a degenerate point mass at , the cumulant generating function is the straight line K_c(t)=ct , and more generally, K_{X+Y}=K_X+K_Y if and only if and are independent and their cumulant generating functions exist; (subindependence and the existence of second moments sufficing to imply independence. ) The natural exponential family of a distribution may be realized by shifting or translating , and adjusting it vertically so that it always passes through the origin: if is the pdf with cumulant generating function K(t)=\log M(t), and f|\theta is its natural exponential family, then f(x\mid\theta)=\frac1{M(\theta)}e^{\theta x} f(x), and K(t\mid\theta)=K(t+\theta)-K(\theta).
If is finite for a range then if then is analytic and infinitely differentiable for . Moreover for real and is strictly convex, and is strictly increasing.
M(t) = 1+\sum_{n=1}^\infty \frac{\mu'_n t^n}{n!} = \exp \left(\sum_{n=1}^\infty \frac{\kappa_n t^n}{n!}\right) = \exp(K(t)). So the cumulant generating function is the logarithm of the moment generating function K(t) = \log M(t). The first cumulant is the expected value; the second and third cumulants are respectively the second and third central moments (the second central moment is the variance); but the higher cumulants are neither moments nor central moments, but rather more complicated polynomial functions of the moments. The moments can be recovered in terms of cumulants by evaluating the th derivative of \exp(K(t)) at , \mu'_n = M^{(n)}(0) = \left. \frac{\mathrm{d}^n \exp (K(t))}{\mathrm{d}t^n}\right|_{t=0}.
Likewise, the cumulants can be recovered in terms of moments by evaluating the th derivative of \log M(t) at , \kappa_n = K^{(n)}(0) = \left. \frac{\mathrm{d}^n \log M(t)}{\mathrm{d}t^n} \right|_{t=0}. The explicit expression for the th moment in terms of the first cumulants, and vice versa, can be obtained by using Faà di Bruno's formula for higher derivatives of composite functions.
In general, we have \mu'_n = \sum_{k=1}^n B_{n,k}(\kappa_1,\ldots,\kappa_{n-k+1}) \kappa_n = \sum_{k=1}^n (-1)^{k-1} (k-1)! B_{n,k}(\mu'_1, \ldots, \mu'_{n-k+1}), where B_{n,k} are incomplete (or partial) Bell polynomials.
In the like manner, if the mean is given by \mu , the central moment generating function is given by C(t) = \operatorname{E}[e^{t(x-\mu)}] = e^{-\mu t} M(t) = \exp(K(t) - \mu t), and the th central moment is obtained in terms of cumulants as \mu_n = C^{(n)}(0) = \left. \frac{\mathrm{d}^n}{\mathrm{d}t^n} \exp (K(t) - \mu t) \right|_{t=0} = \sum_{k=1}^n B_{n,k}(0,\kappa_2,\ldots,\kappa_{n-k+1}). Also, for , the th cumulant in terms of the central moments is \begin{align} \kappa_n & = K^{(n)}(0) = \left. \frac{\mathrm{d}^n}{\mathrm{d}t^n} (\log C(t) + \mu t) \right|_{t=0} \\[4pt] & = \sum_{k=1}^n (-1)^{k-1} (k-1)!
B_{n,k}(0,\mu_2,\ldots,\mu_{n-k+1}). \end{align} The th moment is an th-degree polynomial in the first cumulants. The first few expressions are: \begin{align} \mu'_1 = {} & \kappa_1 \\[5pt] \mu'_2 = {} & \kappa_2+\kappa_1^2 \\[5pt] \mu'_3 = {} & \kappa_3+3\kappa_2\kappa_1+\kappa_1^3 \\[5pt] \mu'_4 = {} & \kappa_4 + 4\kappa_3\kappa_1 + 3\kappa_2^2 + 6\kappa_2\kappa_1^2 + \kappa_1^4 \\[5pt] \mu'_5 = {} & \kappa_5+5\kappa_4\kappa_1+10\kappa_3\kappa_2 + 10\kappa_3\kappa_1^2 + 15\kappa_2^2\kappa_1 + 10\kappa_2\kappa_1^3 + \kappa_1^5 \\[5pt] \mu'_6 = {} & \kappa_6 + 6\kappa_5\kappa_1 + 15\kappa_4\kappa_2 + 15\kappa_4\kappa_1^2 + 10\kappa_3^2 + 60\kappa_3\kappa_2\kappa_1 + 20\kappa_3\kappa_1^3 \\ & {} + 15\kappa_2^3 + 45\kappa_2^2\kappa_1^2 + 15\kappa_2\kappa_1^4 + \kappa_1^6. \end{align} The "prime" distinguishes the moments from the central moments .
To express the central moments as functions of the cumulants, just drop from these polynomials all terms in which appears as a factor: \begin{align} \mu_1 & =0 \\[4pt] \mu_2 & =\kappa_2 \\[4pt] \mu_3 & =\kappa_3 \\[4pt] \mu_4 & =\kappa_4+3\kappa_2^2 \\[4pt] \mu_5 & =\kappa_5+10\kappa_3\kappa_2 \\[4pt] \mu_6 & =\kappa_6+15\kappa_4\kappa_2+10\kappa_3^2+15\kappa_2^3. \end{align} Similarly, the th cumulant is an th-degree polynomial in the first non-central moments. The first few expressions are: \begin{align} \kappa_1 = {} & \mu'_1 \\[4pt] \kappa_2 = {} & \mu'_2-{\mu'_1}^2 \\[4pt] \kappa_3 = {} & \mu'_3-3\mu'_2\mu'_1+2{\mu'_1}^3 \\[4pt] \kappa_4 = {} & \mu'_4-4\mu'_3\mu'_1-3{\mu'_2}^2+12\mu'_2{\mu'_1}^2-6{\mu'_1}^4 \\[4pt] \kappa_5 = {} & \mu'_5-5\mu'_4\mu'_1-10\mu'_3\mu'_2 + 20\mu'_3{\mu'_1}^2 + 30{\mu'_2}^2\mu'_1-60\mu'_2{\mu'_1}^3 + 24{\mu'_1}^5 \\[4pt] \kappa_6 = {} & \mu'_6-6\mu'_5\mu'_1-15\mu'_4\mu'_2+30\mu'_4{\mu'_1}^2-10{\mu'_3}^2 + 120\mu'_3\mu'_2\mu'_1 \\ & {} - 120\mu'_3{\mu'_1}^3 + 30{\mu'_2}^3 - 270{\mu'_2}^2 {\mu'_1}^2+360\mu'_2{\mu'_1}^4-120{\mu'_1}^6\,. \end{align} In general, the cumulant is the determinant of a matrix: \kappa_l = (-1)^{l+1} \left|\begin{array}{cccccccc} \mu'_1 & 1 & 0 & 0 & 0 & 0 & \ldots & 0 \\ \mu'_2 & \mu'_1 & 1 & 0 & 0 & 0 & \ldots & 0 \\ \mu'_3 & \mu'_2 & \binom21 \mu'_1 & 1 & 0 & 0 & \ldots & 0 \\ \mu'_4 & \mu'_3 & \binom31 \mu'_2 & \binom32 \mu'_1 & 1 & 0 & \ldots & 0 \\ \mu'_5 & \mu'_4 & \binom41 \mu'_3 & \binom42 \mu'_2 & \binom43 \mu'_1 & 1 & \ldots & 0 \\ \vdots & \vdots & \vdots & \vdots & \vdots & \ddots & \ddots & \vdots \\ \mu'_{l-1} & \mu'_{l-2} & \ldots & \ldots & \ldots & \ldots & \ddots & 1 \\ \mu'_l & \mu'_{l-1} & \ldots & \ldots & \ldots & \ldots & \ldots & \binom{l-1}{l-2} \mu'_1 \end{array}\right| To express the cumulants for as functions of the central moments, drop from these polynomials all terms in which μ' 1 appears as a factor: \kappa_2=\mu_2\, \kappa_3=\mu_3\, \kappa_4=\mu_4-3{\mu_2}^2\, \kappa_5=\mu_5-10\mu_3\mu_2\, \kappa_6=\mu_6-15\mu_4\mu_2-10{\mu_3}^2+30{\mu_2}^3\,.
The cumulants can be related to the moments by differentiating the relationship with respect to , giving , which conveniently contains no exponentials or logarithms. Equating the coefficient of on the left and right sides and using gives the following formulas for : \begin{align} \mu'_1 = {} & \kappa_1 \\[1pt] \mu'_2 = {} & \kappa_1\mu'_1+\kappa_2 \\[1pt] \mu'_3 = {} & \kappa_1\mu'_2+2\kappa_2\mu'_1+\kappa_3 \\[1pt] \mu'_4 = {} & \kappa_1\mu'_3+3\kappa_2\mu'_2+3\kappa_3\mu'_1+\kappa_4 \\[1pt] \mu'_5 = {} & \kappa_1\mu'_4+4\kappa_2\mu'_3+6\kappa_3\mu'_2+4\kappa_4\mu'_1+\kappa_5 \\[1pt] \mu'_6 = {} & \kappa_1\mu'_5+5\kappa_2\mu'_4+10\kappa_3\mu'_3+10\kappa_4\mu'_2+5\kappa_5\mu'_1+\kappa_6 \\[1pt] \mu'_n = {} & \sum_{m=1}^{n-1}{n-1 \choose m-1}\kappa_m \mu'_{n-m} + \kappa_n\,. \end{align} These allow either \kappa_n or \mu'_n to be computed from the other using knowledge of the lower-order cumulants and moments.
The corresponding formulas for the central moments \mu_n for n \ge 2 are formed from these formulas by setting \mu'_1 = \kappa_1 = 0 and replacing each \mu'_n with \mu_n for n \ge 2 : \begin{align} \mu_2 = {} & \kappa_2 \\[1pt] \mu_3 = {} & \kappa_3 \\[1pt] \mu_n = {} & \sum_{m=2}^{n-2}{n-1 \choose m-1}\kappa_m \mu_{n-m} + \kappa_n\,. \end{align} \mu'_n=\sum_{\pi \, \in \, \Pi} \prod_{B \, \in \, \pi} \kappa_ where runs through the list of all partitions of a set of size ; " " means is one of the "blocks" into which the set is partitioned; and is the size of the set . Thus each monomial is a constant times a product of cumulants in which the sum of the indices is (e.g., in the term , the sum of the indices is 3 + 2 + 2 + 1 = 8; this appears in the polynomial that expresses the 8th moment as a function of the first eight cumulants).
A partition of the integer corresponds to each term. The coefficient in each term is the number of partitions of a set of members that collapse to that partition of the integer when the members of the set become indistinguishable.
Joint cumulants
The joint cumulant of several random variables is defined as the coefficient in the Maclaurin series of the multivariate cumulant generating function, see Section 3.1 in, G(t_1,\dots,t_n)=\log \mathrm{E}(\mathrm{e}^{\sum_{j=1}^n t_j X_j}) =\sum_{k_1,\ldots,k_n} \kappa_{k_1,\ldots,k_n} \frac{t_1^{k_1} \cdots t_n^{k_n}}{k_1! \cdots k_n!} \,. Note that \kappa_{k_1,\dots,k_n} = \left.\left(\frac{\mathrm{d}}{\mathrm{d} t_1}\right)^{k_1} \cdots \left(\frac{\mathrm{d}}{\mathrm{d} t_n}\right)^{k_n} G(t_1,\dots,t_n) \right|_{t_1 = \dots = t_n = 0}\,, and, in particular \kappa(X_1,\ldots,X_n) = \left. \frac{\mathrm{d}^n}{\mathrm{d} t_1 \cdots \mathrm{d} t_n} G(t_1,\dots,t_n) \right|_{t_1 = \dots = t_n = 0}\,.
As with a single variable, the generating function and cumulant can instead be defined via H(t_1,\dots,t_n) =\log \mathrm{E}(\mathrm{e}^{\sum_{j=1}^n i t_j X_j}) =\sum_{k_1,\ldots,k_n} \kappa_{k_1,\ldots,k_n} i^{k_1+\cdots+k_n} \frac{t_1^{k_1} \cdots t_n^{k_n}}{k_1! \cdots k_n!}\,, in which case \kappa_{k_1,\dots,k_n} = (-i)^{k_1+\cdots+k_n} \left.\left(\frac{\mathrm{d}}{\mathrm{d} t_1}\right)^{k_1} \cdots \left(\frac{\mathrm{d}}{\mathrm{d} t_n}\right)^{k_n} H(t_1,\dots,t_n) \right|_{t_1 = \dots = t_n = 0}\,, and \kappa(X_1,\ldots,X_n) = \left. (-i)^{n} \frac{\mathrm{d}^n}{\mathrm{d} t_1 \cdots \mathrm{d} t_n} H(t_1,\dots,t_n) \right|_{t_1 = \dots = t_n = 0}\,.
Observe that \kappa_{k_1,\dots,k_n} (X_1,\ldots,X_n) can also be written as \kappa_{k_1,\dots,k_n} = \left. \frac{\mathrm{d}^{k_1}}{\mathrm{d} t_{1,1} \cdots \mathrm{d} t_{1,k_1}} \cdots \frac{\mathrm{d}^{k_n}}{\mathrm{d} t_{n,1} \cdots \mathrm{d} t_{n,k_n}} G\left(\sum_{j=1}^{k_1}t_{1,j},\dots,\sum_{j=1}^{k_n}t_{n,j}\right) \right|_{t_{i,j}=0}, from which we conclude that \kappa_{k_1,\dots,k_n} (X_1,\ldots,X_n) = \kappa_{1,\ldots,1} ( \underbrace{X_1,\dots,X_1}_{k_1}, \ldots , \underbrace{X_n,\dots,X_n}_{k_n} ) . For example \kappa_{2,0,1}(X,Y,Z) = \kappa(X,X,Z),\, and \kappa_{0,0,n,0}(X,Y,Z,T) = \kappa_{n}(Z) = \kappa(\underbrace{Z,\dots,Z}_{n}) .\, In particular, the last equality shows that the cumulants of a single random variable are the joint cumulants of multiple copies of that random variable. \kappa(X_1,\dots,X_n)=\sum_\pi (|\pi|-1)!(-1)^{|\pi|-1}\prod_{B\in\pi}E\left(\prod_{i\in B}X_i\right) where runs through the list of all partitions of ; where runs through the list of all blocks of the partition ; and where is the number of parts in the partition.
For example, \kappa(X)=\operatorname E(X), is the expected value of X , \kappa(X,Y)=\operatorname E(XY) - \operatorname E(X) \operatorname E(Y), is the covariance of X and Y , and \kappa(X,Y,Z)=\operatorname E(XYZ) - \operatorname E(XY) \operatorname E(Z) - \operatorname E(XZ) \operatorname E(Y) - \operatorname E(YZ) \operatorname E(X) + 2\operatorname E(X)\operatorname E(Y)\operatorname E(Z).\, For zero-mean random variables X_1,\ldots,X_n , any mixed moment of the form \prod_{B\in\pi} E\left(\prod_{i\in B} X_i\right) vanishes if \pi is a partition of \{ 1,\ldots,n \} which contains a singleton B=\{k\} . Hence, the expression of their joint cumulant in terms of mixed moments simplifies.
For example, if X,Y,Z,W are zero mean random variables, we have \kappa(X,Y,Z) = \operatorname E(XYZ).\, \kappa(X,Y,Z,W) = \operatorname E(XYZW) - \operatorname E(XY) \operatorname E(ZW) - \operatorname E(XZ) \operatorname E(YW) - \operatorname E(XW) \operatorname E(YZ).\, More generally, any coefficient of the Maclaurin series can also be expressed in terms of mixed moments, although there are no concise formulae. Indeed, as noted above, one can write it as a joint cumulant by repeating random variables appropriately, and then apply the above formula to express it in terms of mixed moments.
For example \kappa_{201}(X,Y,Z) = \kappa(X,X,Z)=\operatorname E(X^2Z) -2\operatorname E(XZ)\operatorname E(X) - \operatorname E(X^2)\operatorname E(Z) + 2\operatorname E(X)^2\operatorname E(Z).\, If some of the random variables are independent of all of the others, then any cumulant involving two (or more) independent random variables is zero. The combinatorial meaning of the expression of mixed moments in terms of cumulants is easier to understand than that of cumulants in terms of mixed moments, see Equation (3.2.6) in: \operatorname E(X_1\cdots X_n)=\sum_\pi\prod_{B\in\pi}\kappa(X_i : i \in B). For example: \operatorname E(XYZ) = \kappa(X,Y,Z) + \kappa(X,Y)\kappa(Z) + \kappa(X,Z)\kappa(Y) + \kappa(Y,Z)\kappa(X) + \kappa(X)\kappa(Y)\kappa(Z).\, \kappa(X+Y,Z_1,Z_2,\dots) = \kappa(X,Z_1,Z_2,\ldots) + \kappa(Y,Z_1,Z_2,\ldots).\, Just as the second cumulant is the variance, the joint cumulant of just two random variables is the covariance.
The familiar identity \operatorname{var}(X+Y) = \operatorname{var}(X) + 2\operatorname{cov}(X,Y) + \operatorname{var}(Y)\, generalizes to cumulants: \kappa_n(X+Y)=\sum_{j=0}^n {n \choose j} \kappa( \, \underbrace{X,\dots,X}_j, \underbrace{Y,\dots,Y}_{n-j}\,).\, \mu_3(X) = \operatorname E(\mu_3(X\mid Y)) + \mu_3(\operatorname E(X\mid Y)) + 3 \operatorname{cov}(\operatorname E(X\mid Y), \operatorname{var} (X\mid Y)). In general, \kappa(X_1,\dots,X_n)=\sum_\pi \kappa(\kappa(X_{\pi_1}\mid Y), \dots, \kappa(X_{\pi_b}\mid Y)) where the sum is over all partitions of the set of indices, and 1 , ..., b are all of the "blocks" of the partition ; the expression indicates that the joint cumulant of the random variables whose indices are in that block of the partition. \kappa_{n+1}(X\mid Y=y) = \frac{ \mathrm{d}^n}{ \mathrm{d} y^n}\operatorname E(X\mid Y = y), \, n \in \mathbb{N}, \, y \in \mathbb{R}. The results can also be extended to the exponential family.
A system in equilibrium with a thermal bath at temperature has a fluctuating internal energy , which can be considered a random variable drawn from a distribution E\sim p(E) . The partition function of the system is Z(\beta) = \sum_i e^{-\beta E_i} , where = and is the Boltzmann constant and the notation \langle A \rangle has been used rather than \operatorname{E}[A] for the expectation value to avoid confusion with the energy, . Hence the first and second cumulant for the energy give the average energy and heat capacity. \begin{align} \langle E \rangle_c & = \frac{\partial \log Z}{\partial (-\beta)} = \langle E \rangle \\[6pt] \langle E^2 \rangle_c & = \frac{\partial\langle E\rangle_c}{\partial (-\beta)} = k T^2 \frac{\partial \langle E\rangle}{\partial T} = kT^2C \end{align} The Helmholtz free energy expressed in terms of F(\beta) = -\beta^{-1}\log Z(\beta) \, further connects thermodynamic quantities with cumulant generating function for the energy.
Thermodynamics properties that are derivatives of the free energy, such as its internal energy, entropy, and specific heat capacity, all can be readily expressed in terms of these cumulants. Other free energies can be a function of other variables such as the magnetic field or chemical potential \mu , e.g. \Omega=-\beta^{-1}\log\left(\left\langle e^{-\beta (E + \mu N)} \right\rangle\right), where is the number of particles and \Omega is the grand potential.
Again the close relationship between the definition of the free energy and the cumulant generating function implies that various derivatives of this free energy can be written in terms of joint cumulants of and . 1+\sum_{n=1}^\infty \frac{m_n t^n}{n!} = \exp \left( \sum_{n=1}^\infty \frac{\kappa_n t^n}{n!} \right) , where the values of for are found formally, i.e., by algebra alone, in disregard of questions of whether any series converges.
All of the difficulties of the "problem of cumulants" are absent when one works formally. The simplest example is that the second cumulant of a probability distribution must always be nonnegative, and is zero only if all of the higher cumulants are zero.
Formal cumulants are subject to no such constraints. \begin{align} \mu'_6 = \kappa_6 & + 6\kappa_5\kappa_1 + 15\kappa_4\kappa_2 + 15\kappa_4\kappa_1^2 + 10\kappa_3^2+60\kappa_3\kappa_2\kappa_1 \\[4pt] & {} + 20\kappa_3\kappa_1^3 + 15\kappa_2^3 + 45\kappa_2^2\kappa_1^2 + 15\kappa_2\kappa_1^4 + \kappa_1^6 \end{align} make a new polynomial in these plus one additional variable : \begin{align} p_6(x) = \kappa_6 \,x {} & + \left(6\kappa_5\kappa_1 + 15\kappa_4\kappa_2 + 10\kappa_3^2\right) x^2 + \left(15\kappa_4\kappa_1^2 + 60\kappa_3\kappa_2\kappa_1 + 15\kappa_2^3\right) x^3 \\ & {} + 45\kappa_2^2\kappa_1^2 \, x^4 + 15\kappa_2\kappa_1^4 \, x^5 + \kappa_1^6 \, x^6, \end{align} and then generalize the pattern. The pattern is that the numbers of blocks in the aforementioned partitions are the exponents on .
Each coefficient is a polynomial in the cumulants; these are the Bell polynomials, named after Eric Temple Bell. This sequence of polynomials is of binomial type.
In fact, no other sequences of binomial type exist; every polynomial sequence of binomial type is completely determined by its sequence of formal cumulants. \operatorname E(X_1\cdots X_n)=\sum_\pi\prod_{B\,\in\,\pi}\kappa(X_i : i\in B) for joint cumulants, one sums over all partitions of the set . If instead, one sums only over the noncrossing partitions, then, by solving these formulae for the \kappa in terms of the moments, one gets free cumulants rather than conventional cumulants treated above.
These free cumulants were introduced by Roland Speicher and play a central role in free probability theory. In that theory, rather than considering independence of random variables, defined in terms of tensor products of algebras of random variables, one considers instead free independence of random variables, defined in terms of free products of algebras.
The ordinary cumulants of degree higher than 2 of the normal distribution are zero. The free cumulants of degree higher than 2 of the Wigner semicircle distribution are zero. This is one respect in which the role of the Wigner distribution in free probability theory is analogous to that of the normal distribution in conventional probability theory.
Text derived from https://en.wikipedia.org/w/index.php?oldid=1378013955 under CC BY-SA 4.0; modifications have been made.