Moment-generating function ($M_X(t)$)

The moment-generating function is $M_X(t)=E[e^{tX}]$ where finite and encodes moments through derivatives at zero.

Suppose a positive function $M$ satisfies the displayed fourth-power identity and $\log M(h)=h^2/2+o(h^2)$ at zero. Iteration gives $\log M(t)=4^m\log M(t/2^m)$, which tends to $t^2/2$. A related first-power symmetry step is: if $\psi (t)=\psi (t/2{)}^2$ and $\psi (h)=1+o(h^2)$, then $\log \psi (t)=2^m\log \psi (t/2^m)\to 0$. These two different dyadic exponents must not be interchanged in the Gaussian characterization by independent sum and difference.

If $X\ge 0$ and $\mathbb{E}{e}^{tX}<\infty$ for some finite $t\ge 0$, the estimator ${\hat{m}}_n(s)=n^{-1}{\sum }_i{e}^{s{X}_i}$ satisfies ${\sup }_{s\in [0,t]}|{\hat{m}}_n(s)-\mathbb{E}{e}^{sX}|\to 0$ almost surely. The dominated convergence theorem makes the population moment-generating function uniformly continuous on this interval. Parameter subintervals give function brackets whose endpoints are ordered exponentials and whose widths are the corresponding increments of the population mean. Apply the uniform strong law from finite L1 bracketing. Continuity in the parameter makes the supremum measurable.

If two random vectors have moment-generating functions that are finite and equal on a neighborhood of the origin, then they have the same probability distribution.

An exponential moment of a random variable $X$ is an expectation $\mathbb{E}[e^{\theta X}]$ or $\mathbb{E}[e^{\theta |X|}]$ at a specified parameter $\theta$. Its finiteness controls tail decay and the domain of the moment-generating function.

The cumulant-generating function is ${\psi }_X(\theta )=\log \mathbb{E}{e}^{\theta X}$, interpreted as an extended-real convex function where the exponential moment may be infinite.

The $n$th cumulant is the coefficient of $t^n/n!$ in the cumulant-generating function. The first two cumulants are the expected value and variance; the third is the third central moment.

The cumulant expansion writeswhere ${\kappa }_n$ is the $n$th cumulant. Applied to a fluctuating interaction, it expands the logarithm of the averaged Boltzmann weight in connected expectation values.

An Edgeworth expansion approximates a probability density function by a normal distribution multiplied by Hermite polynomials encoding higher cumulants. For a centered variable of variance ${\sigma }^2$ and third cumulant ${\kappa }_3$,The correction follows from inverse-transforming the cubic term of the cumulant-generating function. It is an asymptotic approximation near the bulk of the distribution; truncation can give a negative result in far tails.

The Legendre transform is the convex function

For a uniform random sign $X\in \{-1,1\}$, $\psi (\lambda )=\log \cosh \lambda$ andfor $|x|\le 1$, with value $+\infty$ outside that interval.

For independent and identically distributed random variables with a moment-generating function finite on an open interval containing zero, the empirical mean satisfies a large deviation principle with rate function given by the Legendre transform of a cumulant-generating function. The exponential-moment hypothesis is part of this theorem; mere integrability does not guarantee this full large-deviation principle.

For independent unit-rate exponential random variables, the cumulant-generating function is $-\log (1-\theta )$ for $\theta <1$. Its Legendre transform of a cumulant-generating function is $I(x)=x-1-\log x$ for $x>0$ and infinite otherwise. The Cramér theorem gives the upper-tail logarithmic rate $-I(a)$ for $a>1$, and zero for $0\le a\le 1$. At the mean threshold the infima on both the open and closed upper half-lines are zero; below the mean the strong law of large numbers applies.

For independent identically distributed real variables with everywhere finite cumulant-generating function, let $\mu$ be the mean. If $a\ge \mu$ and $\mathbb{P}(X_1>a)>0$, then $\mathbb{P}(S_n/n\in [a,a+\epsilon )\mid S_n/n\ge a)\to 1$ for each $\epsilon >0$. The Legendre transform of a cumulant-generating function is finite at $a$, and a positive exponential parameter gives strictly greater cost at $a+\epsilon$. A Chernoff bound for the farther tail divided by the Cramér theorem lower estimate at $a$ decays exponentially. The support condition both ensures positive conditioning probabilities and avoids a vacuous conditioning event.

Exponential tilting by $\theta$ replaces a law $P$ byUnder the tilted law, the mean is ${\psi }^{\prime }(\theta )$ whenever the derivative exists.