Uniform Distribution (Continuous)

Uniform distribution (continuous)
In probability theory and statistics, the continuous uni

form distribution or rectangular distribution is a fam
0
forx < a
ily of symmetric probability distributions such that for
each member of the family, all intervals of the same

F (x) = xa
fora x b
ba
length on the distributions support are equally probable.
The support is dened by the two parameters, a and b,

1
forx > b
which are its minimum and maximum values. The distribution is often abbreviated U(a,b). It is the maximum Its inverse is:
entropy probability distribution for a random variate X
under no constraint other than that it is contained in the
distributions support.[1]
F 1 (p) = a + p(b a) for 0 < p < 1
1
1.1
In mean and variance notation, the cumulative distribution function is:
Characterization
0 (
Probability density function
The probability density function of the continuous uni- F (x) = 12
form distribution is:

1
f (x) =
1
ba
for a x b,
for x < a or x > b
)
+1
forx < 3
for 3 x < 3
forx 3
and the inverse is:
F 1 (p) = 3(2p 1) + for 0 p 1
The values of f(x) at the two boundaries a and b are usually unimportant because they do not alter the values of
the integrals of f(x) dx over any interval, nor of x f(x)
dx or any higher moment. Sometimes they are chosen
to be zero, and sometimes chosen to be 1/(b a). The
latter is appropriate in the context of estimation by the
method of maximum likelihood. In the context of Fourier
analysis, one may take the value of f(a) or f(b) to be
1/(2(b a)), since then the inverse transform of many
integral transforms of this uniform function will yield
back the function itself, rather than a function which is
equal "almost everywhere", i.e. except on a set of points
with zero measure. Also, it is consistent with the sign
function which has no such ambiguity.
1.3 Generating functions

1.3.1 Moment-generating function
The moment-generating function is:[2]
Mx = E(etx ) =
etb eta
t(b a)
from which we may calculate the raw moments m k
a+b
In terms of mean and variance 2 , the probability den,
m1 =
sity may be written as:
2
{
f (x) =
1.2
1
2 3
for 3 x 3
otherwise
m2 =
a2 + ab + b2
,
3
mk =
1 i ki
ab .
k + 1 i=0
k
Cumulative distribution function
For a random variable following this distribution, the

expected value is then m1 = (a + b)/2 and the variance
is m2 m1 2 = (b a)2 /12.
The cumulative distribution function is:

1
2
1.3.2
RELATED DISTRIBUTIONS
Cumulant-generating function
x+d
dy
d
For n 2, the nth cumulant of the uniform distribution
P (X [x, x + d]) =
=
on the interval [1/2, 1/2] is bn/n, where bn is the nth
b
a
b
a
x
Bernoulli number.[3]
which is independent of x. This fact motivates the distributions name.
2
2.1
Properties
Moments and parameters
The rst moment of the distribution is:
E(X) =
1
(a + b).
2
The second centralized moment (or variance) is:
V (X) =
1
(b a)2
12
2.4 Generalization to Borel sets

This distribution can be generalized to more complicated
sets than intervals. If S is a Borel set of positive, nite
measure, the uniform probability distribution on S can
be specied by dening the pdf to be zero outside S and
constantly equal to 1/K on S, where K is the Lebesgue
measure of S.
3 Standard uniform
Solving these two equations for parameters a and b, given Restricting a = 0 and b = 1 , the resulting distribution
known moments E(X) and V(X), yields:
U(0,1) is called a standard uniform distribution.
a = E(X) 3V (X)
b = E(X) + 3V (X)
2.2
Order statistics
Let X1 , ..., Xn be an i.i.d. sample from U(0,1). Let Xk

be the kth order statistic from this sample. Then the probability distribution of Xk is a Beta distribution with parameters k and n k + 1. The expected value is
E(X(k) ) =
k
.
n+1
This fact is useful when making Q-Q plots.

The variances are
V(X(k) ) =
2.3
k(n k + 1)
.
(n + 1)2 (n + 2)
Uniformity
The probability that a uniformly distributed random variable falls within any interval of xed length is independent
of the location of the interval itself (but it is dependent on
the interval size), so long as the interval is contained in the
distributions support.
To see this, if X ~ U(a,b) and [x, x+d] is a subinterval of
[a,b] with xed d > 0, then
One interesting property of the standard uniform distribution is that if u1 has a standard uniform distribution,
then so does 1-u1 . This property can be used for generating antithetic variates, among other things.
4 Related distributions
If X has a standard uniform distribution, then by the
inverse transform sampling method, Y = 1 ln(X)
has an exponential distribution with (rate) parameter
.
If X has a standard uniform distribution, then Y =
Xn has a beta distribution with parameters 1/n and
1. (Note this implies that the standard uniform distribution is a special case of the beta distribution,
with parameters 1 and 1.)
The IrwinHall distribution is the sum of n i.i.d.
U(0,1) distributions.
The sum of two independent, equally distributed,
uniform distributions yields a symmetric triangular
distribution.
The distance between two i.i.d. uniform random
variables also has a triangular distribution, although
not symmetric.
The uniform distribution can be thought of as a beta
distribution with parameters (1,1).
6.3
Quantization error
Relationship to other functions
3
for the cases where the cdf is not known in closed form.
One such method is rejection sampling.
As long as the same conventions are followed at the tran- The normal distribution is an important example where
sition points, the probability density function may also be the inverse transform method is not ecient. However,
expressed in terms of the Heaviside step function:
there is an exact method, the BoxMuller transformation,
which uses the inverse transform to convert two independent uniform random variables into two independent
H(x a) H(x b)
normally distributed random variables.
f (x) =
,
ba
or in terms of the rectangle function
1
rect
f (x) =
ba
(
))
x a+b
2
.
ba
There is no ambiguity at the transition point of the sign

function. Using the half-maximum convention at the
transition points, the uniform distribution may be expressed in terms of the sign function as:
f (x) =
sgn (x a) sgn (x b)
.
2(b a)
Applications
6.3 Quantization error

Main article: Quantization error
In analog-to-digital conversion a quantization error occurs. This error is either due to rounding or truncation.
When the original signal is much larger than one least
signicant bit (LSB), the quantization error is not signicantly correlated with the signal, and has an approximately uniform distribution. The RMS error therefore
follows from the variance of this distribution.
7 Estimation
7.1 Estimation of maximum
In statistics, when a p-value is used as a test statistic for 7.1.1 Minimum-variance unbiased estimator
a simple null hypothesis, and the distribution of the test
statistic is continuous, then the p-value is uniformly dis- Main article: German tank problem
tributed between 0 and 1 if the null hypothesis is true.
6.1
Sampling from a uniform distribution
Given a uniform distribution on [0, b] with unknown b,

the minimum-variance unbiased estimator (UMVU) estimator for the maximum is given by
There are many applications in which it is useful to

run simulation experiments. Many programming lanm
k+1
m=m+
guages have the ability to generate pseudo-random num- bU M V U =
k
k
bers which are eectively distributed according to the
where m is the sample maximum and k is the sample size,
standard uniform distribution.
sampling without replacement (though this distinction alIf u is a value sampled from the standard uniform distri- most surely makes no dierence for a continuous distribution, then the value a + (b a)u follows the uniform bution). This follows for the same reasons as estimation
distribution parametrised by a and b, as described above. for the discrete distribution, and can be seen as a very
simple case of maximum spacing estimation. This prob6.2 Sampling from an arbitrary distribu- lem is commonly known as the German tank problem,
due to application of maximum estimation to estimates
tion
of German tank production during World War II.
Main article: Inverse transform sampling
7.1.2 Maximum Likelihood estimator
The uniform distribution is useful for sampling from arbitrary distributions. A general method is the inverse transform sampling method, which uses the cumulative distribution function (CDF) of the target random variable.
This method is very useful in theoretical work. Since simulations using this method require inverting the CDF of
the target variable, alternative methods have been devised
The maximum likelihood estimator is given by:

bM L = m
where m is the sample maximum, also denoted as m =
X(n) the maximum order statistic of the sample.
11
7.1.3
Method of moment estimator
EXTERNAL LINKS
[2] Casella & Berger 2001, p. 626
The method of moments estimator is given by:
[3] https://galton.uchicago.edu/~{}wichura/Stat304/
Handouts/L18.cumulants.pdf
bM M = 2X
[4] Nechval KN, Nechval NA, Vasermanis EK, Makeev VY

(2002) Constructing shortest-length condence intervals.
Transport and Telecommunication 3 (1) 95-103
is the sample mean.

where X
7.2
10 Further reading
Estimation of midpoint
The midpoint of the distribution (a + b) / 2 is both the

mean and the median of the uniform distribution. Although both the sample mean and the sample median are
unbiased estimators of the midpoint, neither is as ecient
as the sample mid-range, i.e. the arithmetic mean of
the sample maximum and the sample minimum, which
is the UMVU estimator of the midpoint (and also the
maximum likelihood estimate).
7.3
Condence interval for the maximum
Let X1 , X2 , X3 , ..., X be a sample from U( 0, L ) where

L is the population maximum. Then Xn = max( X1 , X2 ,
X3 , ..., Xn ) has the density[4]
fn (X(n) ) = n
1
L
X(n)
L
)n1
=n
n1
X(n)
Ln
, 0 < Xn < L
The condence interval for the estimated population

maximum is then ( Xn, Xn / 1/n ) where 100 ( 1 )% is the condence level sought. In symbols
X(n) L X(n) /1/n
See also
Beta distribution
BoxMuller transform
Probability plot
Q-Q plot
Rectangular function
Uniform distribution (discrete)
References
[1] Park, Sung Y.; Bera, Anil K. (2009).

Maximum entropy autoregressive conditional heteroskedasticity model. Journal of Econometrics (Elsevier) 150 (2):
219230. doi:10.1016/j.jeconom.2008.12.014.
Casella, George; Roger L. Berger (2001), Statistical

Inference (2nd ed.), ISBN 0-534-24312-6, LCCN
2001025794
11 External links
Online calculator of Uniform distribution (continuous)
12
12.1
Text and image sources, contributors, and licenses

Text
Uniform distribution (continuous) Source: https://en.wikipedia.org/wiki/Uniform_distribution_(continuous)?oldid=693345665 Contributors: Zundark, Gareth Owen, Michael Hardy, Jitse Niesen, Robbot, Robbyjo~enwiki, Henrygb, Giftlite, ArnoldReinhold, Bender235,
MisterSheik, Nsaa, B k, Sl, PAR, Fasten, Aegis Maelstrom, Cburnett, Oleg Alexandrov, Btyner, Marudubshinki, Pahan~enwiki, Mathbot,
Benlisquare, Algebraist, Ray Chason, Ossska, SmackBot, Gilliam, Amatulic, Nbarth, Iwaterpolo, Gala.martin, Euchiasmus, Stwalkerster,
Gritzko, Abdullah Chougle, FilipeS, MC10, AlekseyP, Thijs!bot, Iae, JAnDbot, A.M.R., Albmont, Baccyak4H, User A1, Ceancata, R'n'B,
VolkovBot, DrMicro, Broadbot, Jamelan, Brianga, Tpb, Rlendog, DaBler, DixonD, Melcombe, MATThematical, DragonBot, EtudiantEco,
Sun Creator, Qwfp, Addbot, Brumski, Fgnievinski, Sandrarossi, Wikomidia, Luckas-bot, Yobot, AnomieBOT, Xqbot, Capricorn42, Ruy
Pugliesi, GrouchoBot, Erik9bot, Thehelpfulbot, D'ohBot, Warriorman21, Jonesey95, Stpasha, MastiBot, Duoduoduo, Marie Poise, EmausBot, Vilietha, Tolly4bolly, ClueBot NG, Arthurprat, Ryan Vesey, BG19bot, Ren Vpenk, IkamusumeFan, SFK2, Smirglis, OracleMachine, YiFeiBot, BeyondNormality, EricTheGreen, XenonFlight, Nostroamio, Michalove and Anonymous: 81
12.2
Images
File:Uniform_Distribution_PDF_SVG.svg Source: https://upload.wikimedia.org/wikipedia/commons/9/96/Uniform_Distribution_

PDF_SVG.svg License: CC BY-SA 3.0 Contributors: Own work Original artist: IkamusumeFan
File:Uniform_cdf.svg Source: https://upload.wikimedia.org/wikipedia/commons/6/63/Uniform_cdf.svg License: CC BY-SA 3.0 Contributors: Own work Original artist: IkamusumeFan
12.3
Content license
Creative Commons Attribution-Share Alike 3.0

Uniform Distribution (Continuous)

Uploaded by

Copyright:

Available Formats

Uniform Distribution (Continuous)

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Uniform Distribution (Continuous)

Uploaded by

Copyright:

Available Formats

Uniform distribution (continuous)

In probability theory and statistics, the continuous uni

ily of symmetric probability distributions such that for

each member of the family, all intervals of the same

The support is dened by the two parameters, a and b,

In mean and variance notation, the cumulative distribution function is:

Probability density function

The probability density function of the continuous uni- F (x) = 12

form distribution is:

for x < a or x > b

and the inverse is:

F 1 (p) = 3(2p 1) + for 0 p 1

1.3 Generating functions

from which we may calculate the raw moments m k

Cumulative distribution function

For a random variable following this distribution, the

The cumulative distribution function is:

The rst moment of the distribution is:

The second centralized moment (or variance) is:

2.4 Generalization to Borel sets

Let X1 , ..., Xn be an i.i.d. sample from U(0,1). Let Xk

This fact is useful when making Q-Q plots.

Relationship to other functions

There is no ambiguity at the transition point of the sign

6.3 Quantization error

Sampling from a uniform distribution

Given a uniform distribution on [0, b] with unknown b,

There are many applications in which it is useful to

The maximum likelihood estimator is given by:

Method of moment estimator

[2] Casella & Berger 2001, p. 626

The method of moments estimator is given by:

[4] Nechval KN, Nechval NA, Vasermanis EK, Makeev VY

is the sample mean.

The midpoint of the distribution (a + b) / 2 is both the

Condence interval for the maximum

Let X1 , X2 , X3 , ..., X be a sample from U( 0, L ) where

The condence interval for the estimated population

X(n) L X(n) /1/n

[1] Park, Sung Y.; Bera, Anil K. (2009).

Casella, George; Roger L. Berger (2001), Statistical

Text and image sources, contributors, and licenses

File:Uniform_Distribution_PDF_SVG.svg Source: https://upload.wikimedia.org/wikipedia/commons/9/96/Uniform_Distribution_

Creative Commons Attribution-Share Alike 3.0

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.