info prev up next book cdrom email home

Cumulant-Generating Function

Let $M(h)$ be the Moment-Generating Function. Then

\begin{displaymath}
K(h)\equiv\ln M(h) =\kappa_1 h+{\textstyle{1\over 2!}} h^2\kappa_2+{\textstyle{1\over 3!}} h^3\kappa_3+\ldots.
\end{displaymath}

If

\begin{displaymath}
L=\sum_{j=1}^M c_jx_j
\end{displaymath}

is a function of $N$ independent variables, the cumulant generating function for $L$ is then

\begin{displaymath}
K(h)=\sum_{j=1}^N K_j(c_j h).
\end{displaymath}

See also Cumulant, Moment-Generating Function


References

Abramowitz, M. and Stegun, C. A. (Eds.). Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables, 9th printing. New York: Dover, p. 928, 1972.

Kenney, J. F. and Keeping, E. S. ``Cumulants and the Cumulant-Generating Function'' and ``Additive Property of Cumulants.'' §4.10-4.11 in Mathematics of Statistics, Pt. 2, 2nd ed. Princeton, NJ: Van Nostrand, pp. 77-80, 1951.




© 1996-9 Eric W. Weisstein
1999-05-25