Bispectrum 2026-10-14
The bispectrum is the Fourier transform of a third-order cumulant or correlation function, with the convention specified for the process or field. It measures third-order dependence among frequency triples; the primordial bispectrum is a cosmological random-field application.
Compound binomial distribution Created 2026-10-14 Updated 2026-11-05
For an independent binomial distribution count , the aggregate can also be represented as , with independent parameter- Bernoulli random variables and independent claim sizes. Its cumulant-generating function is . Its third cumulant is in severity moments. Constant positive claims give , which is negative when . Thus positive claim sizes do not by themselves force a positive third central moment.
Compound geometric distribution Created 2026-10-14 Updated 2026-11-05
Take an independent geometric distribution count with , , and sum its independent positive claims. Its moment-generating function is the displayed composition wherever finite. With and severity moments , the third cumulant is , which is strictly positive for nonzero positive claims and . The law has an atom of mass at zero. Its positive-count convention would instead add a factor to the numerator, as in the geometric-sum moment-generating function.
Conditioning on the independent Poisson distribution count in a random sum gives wherever finite. Differentiation at zero gives the displayed cumulants when the relevant derivatives exist. The aggregate variance is the count rate times the raw second claim moment, not just the claim variance. Positive moment-generating function arguments need an exponential moment; finite first and second claim moments suffice for the aggregate mean and variance by conditioning or right derivatives at zero of the Laplace transform of a nonnegative random variable.
For driven by centered unit-variance iid noise with a finite fourth moment, put and . Independence gives
The sum of the two linear-quadratic cross covariances is . The cumulant term disappears for Gaussian noise. Merely assuming strong white noise does not justify the Gaussian formula; a fourth moment is needed for the variance of the quadratic transform.
Cumulant Created 2026-10-05 Updated 2026-11-05
The th cumulant is the coefficient of in the cumulant-generating function. The first two cumulants are the expected value and variance; the third is the third central moment.
Cumulant composition for a random sum Created 2026-10-14 Updated 2026-11-05
For a random sum of independent claims, conditioning on the count gives and therefore the displayed cumulant-generating function. Write and . The chain rule gives , , and . In raw severity moments , the third cumulant is . Positive exponential moments justify two-sided differentiation near zero; finite ordinary moments alone suffice for the corresponding left derivatives when claims are positive.
Cumulant expansion Created 2026-09-24 Updated 2026-11-05
The cumulant expansion is the Taylor expansion of the logarithm of a moment-generating function:
If the moment-generating function is finite on a neighbourhood of zero, the logarithm is analytic on a sufficiently small neighbourhood and this is a convergent local expansion. The first terms are
A formal series of moments is not automatically a convergent representation of a moment-generating function. Finitely many moments can instead justify finite-order expansions of the characteristic function and its logarithm near zero; exponential integrability is a separate hypothesis.
For a centered normal distribution of variance , , so all cumulants beyond order two vanish. For a Poisson distribution of mean , , so every cumulant equals . independent sums add their cumulant-generating functions, making cumulants additive.
For a fluctuating interaction , expanding a Boltzmann weight gives whenever the exponential moment exists near . The logarithm thus organizes the expansion into cumulants rather than raw moments.
Cumulant expansion of a coarse-grained free energy Created 2026-10-05 Updated 2026-11-05
A cumulant expansion expresses the free energy obtained by integrating out fast Fourier modes in terms of their connected fluctuations. Write the physical free energy as , where the quadratic reference part defines a normalized Gaussian measure for the fast modes. At fixed slow field, the interaction contribution is
Here is temperature, is the Boltzmann constant and averages only the fast modes. Expanding the logarithm gives
Every term has units of energy. Equivalently, define the dimensionless interaction and dimensionless free energy ; the expansion then has coefficients without explicit factors. This is the convention commonly used in statistical field theory. [ref]
The subtraction in the second cumulant removes products of disconnected averages. For a Gaussian measure, Wick theorem turns the remaining terms into connected Feynman diagrams. Truncating this formal expansion is a perturbation theory approximation, whose accuracy depends on the interaction and the fluctuation scales.
Edgeworth series Created 2026-10-05 Updated 2026-11-05
An Edgeworth expansion refines the central limit theorem by adding corrections determined by higher cumulants. Let be independent and identically distributed with mean zero, variance and skewness . For the standardized sum , the first density correction is
Here is the standard normal density. This is an asymptotic statement for a sequence, not a universal density formula for one arbitrary random variable. Sufficient smoothness and moment conditions are needed; one strong sufficient setting is that the common probability density function is a non-degenerate Schwartz function. In that setting the displayed remainder is valid for each fixed . Lattice-valued variables need additional corrections rather than this density formula.[ref]
To identify the coefficient, expand the log characteristic function of the standardized sum:
for fixed . Exponentiation and Fourier inversion turn into the Hermite polynomial times . The smoothness assumptions justify inversion beyond this formal calculation. Higher-order terms use further cumulants and Hermite polynomials.
The approximation improves central asymptotics, but it need not be a nonnegative normalized density after truncation. In particular the cubic correction can become negative in a far tail, where a fixed-point expansion does not provide a small relative error. Inverting a distribution-function Edgeworth expansion produces the Cornish-Fisher expansion for fixed quantiles.
Fredholm determinant 2026-10-14
For a trace-class operator on a Hilbert space, the Fredholm determinant is the convergent product over its eigenvalues, counted with algebraic multiplicity. For self-adjoint , spectral theorem for compact self-adjoint operators reduces the product to a diagonal calculation. For , . This identity exposes the successive cumulants in a Gaussian quadratic exponential moment.
For with , insert its probability mass function into the expected value:
The exponential series shows finiteness for every real . Differentiating this moment-generating function gives mean and variance both . Taking the logarithm gives the cumulant-generating function , so every positive-order cumulant is also .
One-loop shell quartic renormalization Created 2026-10-05 Updated 2026-11-05
For centered fast Fourier modes and , the two-fast-field vertex is . Its connected second cumulant uses , producing . Extracting the local zero-external-momentum quartic coefficient gives , where . At its sign makes weak positive quartic coupling marginally irrelevant for increasing length scale, after tuning the critical mass. This local truncation omits generated derivative and higher-order operators; it is not the exact complete finite-shell effective action.
Where the moment-generating function is finite in a neighborhood of zero, define the cumulant-generating function and the th cumulant by . Differentiation gives
since . Thus the first two cumulants are the mean and variance.
Conditional on , independence gives . Summing over the Poisson distribution yields
Hence the compound Poisson cumulants give
For the moment statements, finite second moments suffice even when a moment-generating function does not exist near zero: conditioning gives . This distinguishes the raw second moment from the variance of an individual claim.
For a single flood, the law of total expectation and law of total variance similarly give
Condition on the number of floods and apply the same laws once more. All floods have these moments and are independent of , so the annual total satisfies
The additional term is the variation in the number of entire flood clusters; it would be lost by treating individual claims as an ordinary Poisson stream.
For the annual count , its probability generating function is
This is the nested Poisson flood count. Differentiating, or conditioning on , gives and . Thus is not Poisson when , since its variance exceeds its mean. If either parameter is zero it is identically zero, a degenerate Poisson distribution.
Let be the number of valid claims in flood . Independent validity marking gives . Its generating function is
so Poisson thinning gives . With validity independent also of claim amount, each claim contributes expected paid amount . Therefore the annual expected payment is . The independent-mark assumption is needed for this multiplication; a size-dependent invalidity rule would require the joint expected value of amount and validity instead.
An Edgeworth expansion refines a central normal approximation using higher cumulants. Let be independent and identically distributed, with mean , positive variance , and cumulants . Define
The logarithm of its characteristic function has the expansion
Expanding the exponential adds the squared cubic term . Fourier inversion turns into , where the probabilists' Hermite polynomials satisfy
Thus the density Edgeworth expansion through order is
Here , , and . Since , the corresponding cumulative distribution function expansion is
These remainder statements require adequate moments and smooth nonlattice conditions allowing transform inversion and error control, not merely a formal cumulant expansion. The density version needs the corresponding local smoothness. A symmetric distribution has , removing the first correction. Truncation can produce negative densities or nonmonotone distribution functions far from the center; it is not automatically a uniformly reliable relative tail approximation. A lattice distribution needs its lattice correction.
A Laplace approximation instead approximates an integral dominated by a sharp maximum. Suppose has a unique interior maximizer , with positive definite, and the integral is localized near that point. For smooth amplitude ,
Set . A Taylor expansion gives
Expanding and integrating the resulting Gaussian integral gives
Under sufficient smoothness and localization the relative correction is : odd terms of order integrate to zero over the limiting symmetric Gaussian region. Higher terms use derivatives of and Gaussian moments.
For a likelihood function with and a smooth positive prior density, this gives the integrated likelihood approximation
where is the observed information. Ratios of such integrals give posterior expectation by Laplace approximation; the same method is useful when integrating nuisance variables. Boundary maxima, singular Hessian matrices, nonlocalized tails and several dominant modes require altered formulas, with contributions summed when appropriate.
Both methods start with a quadratic Gaussian approximation and then retain higher-order terms. The Edgeworth series corrects a sampling distribution through its cumulants and Fourier inversion; Laplace's method corrects an integral through local derivatives at its dominant point. Neither should be confused with simply replacing a likelihood by a normalized posterior density.
Use Minkowski spacetime with signature and , with a vacuum time-ordering prescription. At a regulator, define the normalized generating functional
The normalized vacuum correlation functions of a time-ordered product are
The connected generating functional is
The logarithm selects connected source diagrams, or equivalently cumulants. With the assumed vanishing one-point function, derivatives of give the centered four-point cumulant decomposition:
In detail, four differentiations either act on one connected block or form one of three pair partitions. Contributions with blocks of size one vanish. If and denote source derivatives at zero, the same relation is , since and .
Let be the source-dependent mean field. On a locally invertible source-to-field branch, define the quantum effective action by the Minkowski Legendre transform
Its variation is , so . Differentiating the two inverse source-field maps yields the Minkowski inverse-Hessian relation for an effective action:
Thus . The inverse is understood at the regulator and on a nonsingular fluctuation sector; it is not an assertion that every source-field map is globally invertible.
For a free real scalar field, integrate by parts to write
where repeated spacetime variables are integrated. Specify the propagator convention explicitly:
Completing the Gaussian path integral around cancels the source-independent determinant between numerator and denominator. Therefore
The positive imaginary term in damps the oscillatory Gaussian. Together with vacuum projection at the time boundaries, it selects the Feynman poles and the time-ordered, in-out correlation function rather than a retarded inverse. It can also be obtained by continuation from the Euclidean vacuum integral. A denominator convention without the numerator redistributes the displayed factors of .
Finally , hence . Substitution into the Legendre transform gives the vacuum-normalized effective action of a free scalar field:
All connected functions beyond order two vanish and all proper vertices beyond the classical quadratic kernel vanish. Vacuum normalization fixes the additive constant; without it, the quantum effective action equals the classical action up to a field-independent determinant contribution.
Use the Probabilists' Hermite polynomials, rather than the differently scaled physicists' convention:
Thus , , and . The cumulant-generating function of an exponential variable of mean is near zero, so its th cumulant is . Hence the standardized third and fourth cumulants are and ; the fourth is a cumulant, not the fourth raw central moment.
Write for the cumulative distribution function of the pivot in part (i). Its Edgeworth expansion, valid here because the exponential law is smooth, nonlattice and has exponential moments, is
at the fixed endpoints . The coverage is exactly . Since and are even, the terms cancel; the two odd Hermite terms double. Consequently the symmetric Edgeworth coverage cancellation gives
In particular its coverage probability error is . This is a fixed-confidence-level expansion; it is not a uniform statement for normal cutoffs increasing with .
Let . Conditioning on gives
Where the moment-generating function is finite around zero, differentiating the cumulant-generating function gives
In particular and . These compound Poisson cumulants use raw claim moments. The first two formulas also follow by conditioning: and , so
This is the law of total variance. They therefore remain valid with finite first two moments even when no positive exponential moment exists.
Under per-claim excess of loss reinsurance, put and let be the positive part of the excess. The insurer and reinsurer totals are sums of these payments over the same Poisson distribution count. Write . Using the tail integral formula for moments on the tail probabilities of and gives
For example, after , and shifting by gives the reinsurer formula. The moment formulas are finite whenever the corresponding claim moments are finite.
For the specified Pareto distribution, with shape three and minimum ,
If , every claim exceeds the reinsurance retention. Therefore
If , the tail integrals instead give
Consequently the requested sum of variances is
On the first branch, ; it decreases until and then increases. On the second branch . For the minimum is unique; if , all variances are zero and every reinsurance retention minimizes them. Thus
The curve starts at , reaches this minimum, passes through with matching slope on both branches, and then increases concavely toward the horizontal asymptote . The variance-minimizing retention for shape-three Pareto claims is shown below.
Figure 1.
Sum of insurer and reinsurer aggregate variances for shape-three Pareto claims, minimized at retention three quarters of the minimum claim size
. Details.
The two aggregate payouts are dependent. Conditioning on the common count gives covariance , so for every reinsurance retention, as required by . The covariance of two components of a compound Poisson sum accounts for this distinction between a sum of variances and the variance of a sum.
There is a moment domain qualification for this heavy-tailed example. For every , , and moments of order also diverge. Thus its positive moment-generating function is not finite and higher finite cumulants cannot be obtained by differentiating around zero. Its mean and variance calculations above are valid by conditioning or by the first two right derivatives at zero of the Laplace transform of a nonnegative random variable.
Let the summands have mean mu, variance , and cumulants . For the standardized sum , expansion of its log-characteristic function gives a normal leading term and successively smaller cumulant terms. Fourier inversion turns these terms into Probabilists' Hermite polynomials. Under sufficient moments and a smooth nonlattice condition, the Edgeworth expansion begins
for bounded z. Higher orders include the fourth cumulant and products of lower cumulants. This is a central expansion in powers of ; it need not remain positive, normalized, or accurate in far tails after truncation.
For a saddlepoint density approximation, require the summand cumulant-generating function K to be finite on an interval about zero. At a sum value , find the interior saddle satisfying . Exponential tilting makes the tilted mean y, so the tilted sum is approximated locally by its normal probability density at its mean. Undoing the tilt gives
Equivalently this follows by deforming the transform-inversion contour through its stationary point. Suitable analytic and nonlattice regularity gives relative error at fixed interior y. This approximation incorporates a nonquadratic tail exponent and is positive, while an Edgeworth truncation uses a normal probability density times a polynomial. It uses more information from the distribution and stronger exponential-moment assumptions. Its integral may differ from one by a higher-order amount, so positivity does not imply exact normalization.
For the centered Laplace distribution, , hence and . Its skewness vanishes, so the order- Edgeworth correction is zero. With , the requested expansion, omitting terms of order in the standardized correction, is
For bounded z the exact probability density function is this expression times . The prefactor is because this is the probability density function of the unstandardized sum; that change of units should not be confused with the orders inside the Edgeworth expansion.
For the saddlepoint, . Write and . The root in is
These formulas include without a division by y. Substitution yields
This is the leading saddlepoint approximation, without its order- relative correction. Expanding d near one recovers the same central normal approximation, while retaining d captures the Laplace-sum large-deviation exponent away from the center.
Figure 1.
Leading Edgeworth and saddlepoint approximations compared with the exact Laplace-sum density
. Details.
An Edgeworth expansion refines the central limit theorem by incorporating higher cumulants. For iid centred observations with variance , standardized sum , standardized third cumulant and fourth cumulant , its central-region density expansion is
under sufficient moment and smoothness/nonlattice conditions. Here , , are probabilists' Hermite polynomials. Expanding the characteristic exponent gives the cubic and quartic terms, while squaring the cubic term in the exponential gives the correction. Integrating the first correction, using , gives
Thus skewness determines the first departure from the normal approximation; inversion gives a Cornish-Fisher expansion for quantiles. These are asymptotic expansions, not guaranteed nonnegative densities in extreme tails. Lattice data require continuity/lattice corrections, and moment assumptions alone do not justify a uniform smooth-density expansion.