info prev up next book cdrom email home

Laplace Distribution

\begin{figure}\begin{center}\BoxedEPSF{LaplaceDistribution.epsf scaled 650}\end{center}\end{figure}

Also called the Double Exponential Distribution. It is the distribution of differences between two independent variates with identical Exponential Distributions (Abramowitz and Stegun 1972, p. 930).

$\displaystyle P(x)$ $\textstyle =$ $\displaystyle {1\over 2b}e^{-\vert x-\mu\vert/b}$ (1)
$\displaystyle D(x)$ $\textstyle =$ $\displaystyle {\textstyle{1\over 2}}[1+\mathop{\rm sgn}\nolimits (x-\mu)(1-e^{-\vert x-\mu\vert/b})].$ (2)

The Moments about the Mean $\mu_n$ are related to the Moments about 0 by
\begin{displaymath}
\mu_n=\sum_{j=0}^n{n\choose j}(-1)^{n-j}\mu'_j\mu^{n-j},
\end{displaymath} (3)

where ${n\choose k}$ is a Binomial Coefficient, so
$\displaystyle \mu_n$ $\textstyle =$ $\displaystyle \sum_{j=0}^n\sum_{k=0}^{\left\lfloor{j/2}\right\rfloor } (-1)^{n-j}{n\choose j}{j\choose 2k}b^{2k}\mu^{n-2k}\Gamma(2k+1)$  
  $\textstyle =$ $\displaystyle \left\{\begin{array}{ll} n! b^n & \mbox{for $n$\ even}\\  0 & \mbox{for $n$\ odd,}\end{array}\right.$ (4)

where $\left\lfloor{x}\right\rfloor $ is the Floor Function and $\Gamma(2k+1)$ is the Gamma Function.


The Moments can also be computed using the Characteristic Function,

\begin{displaymath}
\phi(t)\equiv\int_{-\infty}^\infty e^{itx} P(x)\,dx
= {1\over 2b} \int_{-\infty}^\infty e^{itx}e^{-\vert x-\mu\vert/b}\,dx.
\end{displaymath} (5)

Using the Fourier Transform of the Exponential Function
\begin{displaymath}
{\mathcal F}[e^{-2\pi k_0\vert x\vert}] = {1\over\pi} {k_0\over k^2+{k_0}^2}
\end{displaymath} (6)

gives
\begin{displaymath}
\phi(t)={e^{i\mu t}\over 2b} {{2\over b}\over t^2+\left({1\over b}\right)^2}
= {e^{i\mu t}\over 1+b^2t^2}.
\end{displaymath} (7)

The Moments are therefore
\begin{displaymath}
\mu_n=(-i)^n\phi(0)=(-i)^n\left[{d^n\phi\over dt^n}\right]_{t=0}.
\end{displaymath} (8)

The Mean, Variance, Skewness, and Kurtosis are
$\displaystyle \mu$ $\textstyle =$ $\displaystyle \mu$ (9)
$\displaystyle \sigma^2$ $\textstyle =$ $\displaystyle 2b^2$ (10)
$\displaystyle \gamma_1$ $\textstyle =$ $\displaystyle 0$ (11)
$\displaystyle \gamma_2$ $\textstyle =$ $\displaystyle 3.$ (12)


References

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



info prev up next book cdrom email home

© 1996-9 Eric W. Weisstein
1999-05-26