Rademacher distribution

From Wikipedia, the free encyclopedia
Rademacher
Support k\in \{-1,1\}\,
pmf f(k)={\begin{cases}1/2,&k=-1\\1/2,&k=1\end{cases}}
CDF F(k)={\begin{cases}0,&k<-1\\1/2,&-1\leq k<1\\1,&k\geq 1\end{cases}}
Mean 0\,
Median 0\,
Mode N/A
Variance 1\,
Skewness 0\,
Ex. kurtosis -2\,
Entropy \ln(2)\,
MGF \cosh(t)\,
CF \cos(t)\,

In probability theory and statistics, the Rademacher distribution (which is named after Hans Rademacher) is a discrete probability distribution where a random variate X has a 50% chance of being either +1 or -1.[1]

A series of Rademacher distributed variables can be regarded as a simple symmetrical random walk where the step size is 1.

Mathematical formulation

The probability mass function of this distribution is

f(k)=\left\{{\begin{matrix}1/2&{\mbox{if }}k=-1,\\1/2&{\mbox{if }}k=+1,\\0&{\mbox{otherwise.}}\end{matrix}}\right.

It can be also written as a probability density function, in terms of the Dirac delta function, as

f(k)={\frac  {1}{2}}\left(\delta \left(k-1\right)+\delta \left(k+1\right)\right).

van Zuijlen's bound

van Zuijlen has proved the following result.[2]

Let Xi be a set of independent Rademacher distributed random variables. Then

\Pr {\Bigl (}{\Bigl |}{\frac  {\sum _{{i=1}}^{n}X_{i}}{{\sqrt  n}}}{\Bigr |}\leq 1)\geq 0.5.

The bound is sharp and better than that which can be derived from the normal distribution (approximately Pr > 0.31).

Bounds on sums

Let { Xi } be a set of random variables with a Rademacher distribution. Let { ai } be a sequence of real numbers. Then

\Pr(\sum _{i}X_{i}a_{i}>t||a_{i}||_{2})\leq e^{{-{\frac  {t^{2}}{2}}}}

where ||ai||2 is the Euclidean norm of the sequence { ai }, t is a real number > 0 and Pr(Z) is the probability of event Z.[3]

Also if ||ai||1 is finite then

\Pr(\sum _{i}X_{i}a_{i}>t||a_{i}||_{1})=0

where || ai ||1 is the 1-norm of the sequence { ai }.

Let Y = Σ Xiai and let Y be an almost surely convergent series in a Banach space. The for t > 0 and s ≥ 1 we have[4]

Pr(||Y||>st)\leq [{\frac  {1}{c}}Pr(||Y||>t)]^{{cs^{2}}}

for some constant c.

Let p be a positive real number. Then[5]

c_{1}[\sum {|a_{i}|^{2}}]^{{\frac  {1}{2}}}\leq (E[|\sum {a_{i}X_{i}}|^{p}])^{{{\frac  {1}{p}}}}\leq c_{2}[\sum {|a_{i}|^{2}}]^{{\frac  {1}{2}}}

where c1 and c2 are constants dependent only on p.

For p ≥ 1

c_{2}\leq c_{1}{\sqrt  {p}}

Applications

The Rademacher distribution has been used in bootstrapping.

The Rademacher distribution can be used to show that normally distributed and uncorrelated does not imply independent.

Related distributions

References

  1. Hitczenko P, Kwapień S (1994) On the Rademacher series. Progress in probability 35: 31-36
  2. van Zuijlen Martien CA (2011) On a conjecture concerning the sum of independent Rademacher random variables. http://arxiv.org/abs/1112.4988
  3. MontgomerySmith SJ (1990) The distribution of Rademacher sums. Proc Amer Math Soc 109: 517522
  4. Dilworth SJ, Montgomery-Smith SJ (1993) The distribution of vector-valued Radmacher series. Ann Probab 21 (4) 2046-2052
  5. Khintchine A (1923) Über dyadische Brüche. Math Zeitschr 18: 109–116

This article is issued from Wikipedia. The text is available under the Creative Commons Attribution/Share Alike; additional terms may apply for the media files.