Raised cosine distribution

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Raised cosine
Probability density function
Plot of the raised cosine PDF
Cumulative distribution function
Plot of the raised cosine CDF


Support x \in [\mu-s,\mu+s]\,
pdf \frac{1}{2s}
CDF \frac{1}{2}\left[1\!+\!\frac{x\!-\!\mu}{s}
Mean \mu\,
Median \mu\,
Mode \mu\,
Variance s^2\left(\frac{1}{3}-\frac{2}{\pi^2}\right)\,
Skewness 0\,
Ex. kurtosis \frac{6(90-\pi^4)}{5(\pi^2-6)^2}\,
MGF \frac{\pi^2\sinh(s t)}{st(\pi^2+s^2 t^2)}\,e^{\mu t}
CF \frac{\pi^2\sin(s t)}{st(\pi^2-s^2 t^2)}\,e^{i\mu t}

In probability theory and statistics, the raised cosine distribution is a continuous probability distribution supported on the interval [\mu-s,\mu+s]. The probability density function is


for \mu-s \le x \le \mu+s and zero otherwise. The cumulative distribution function is


for \mu-s \le x \le \mu+s and zero for x<\mu-s and unity for x>\mu+s.

The moments of the raised cosine distribution are somewhat complicated, but are considerably simplified for the standard raised cosine distribution. The standard raised cosine distribution is just the raised cosine distribution with \mu=0 and s=1. Because the standard raised cosine distribution is an even function, the odd moments are zero. The even moments are given by:

E(x^{2n})=\frac{1}{2}\int_{-1}^1  [1+\cos(x\pi)]x^{2n}\,dx
= \frac{1}{n\!+\!1}+\frac{1}{1\!+\!2n}\,_1F_2

where \,_1F_2 is a generalized hypergeometric function.

Differential equation

\left\{2 s^3 f''(x)-2 \pi ^2 s f(x)+\pi ^2=0,f(0)=\frac{\cosh
   ^2\left(\frac{\pi  \mu }{2 s}\right)}{s},f'(0)=-\frac{\pi  \sinh
   \left(\frac{\pi  \mu }{s}\right)}{2 s^2}\right\}

See also[edit]