» 
Arabic Bulgarian Chinese Croatian Czech Danish Dutch English Estonian Finnish French German Greek Hebrew Hindi Hungarian Icelandic Indonesian Italian Japanese Korean Latvian Lithuanian Malagasy Norwegian Persian Polish Portuguese Romanian Russian Serbian Slovak Slovenian Spanish Swedish Thai Turkish Vietnamese
Arabic Bulgarian Chinese Croatian Czech Danish Dutch English Estonian Finnish French German Greek Hebrew Hindi Hungarian Icelandic Indonesian Italian Japanese Korean Latvian Lithuanian Malagasy Norwegian Persian Polish Portuguese Romanian Russian Serbian Slovak Slovenian Spanish Swedish Thai Turkish Vietnamese

definition - Gamma distribution

definition of Wikipedia

   Advertizing ▼

Wikipedia

Gamma distribution

                   
Gamma
Probability density function
Probability density plots of gamma distributions
Cumulative distribution function
Cumulative distribution plots of gamma distributions
Parameters
Support \scriptstyle x \;\in\; [0,\, \infty)\! \scriptstyle x \;\in\; [0,\, \infty)\!
Probability density function (pdf) \scriptstyle \frac{1}{\Gamma(k) \theta^k} x^{k \,-\, 1} e^{-\frac{x}{\theta}}\,\! \scriptstyle \frac{\beta^\alpha}{\Gamma(\alpha)} x^{\alpha \,-\, 1} e^{-\beta x}\,\!
Cumulative distribution function (cdf) \scriptstyle \frac{1}{\Gamma(k)} \gamma\left(k,\, \frac{x}{\theta}\right)\! \scriptstyle \frac{1}{\Gamma(\alpha)} \gamma(\alpha,\, \beta x)\!
Mean \scriptstyle k \theta \! \scriptstyle \frac{\alpha}{\beta} \!
Median No simple closed form No simple closed form
Mode \scriptstyle (k \,-\, 1)\theta \text{ for } k \;\geq\; 1\,\! \scriptstyle \frac{\alpha \,-\, 1}{\beta} \text{ for } \alpha \;\geq\; 1\,\!
Variance \scriptstyle k \theta^2\,\! \scriptstyle \frac{\alpha}{\beta^2}\,\!
Skewness \scriptstyle \frac{2}{\sqrt{k}}\,\! \scriptstyle \frac{2}{\sqrt{\alpha}}\,\!
Excess kurtosis \scriptstyle \frac{6}{k}\,\! \scriptstyle \frac{6}{\alpha}\,\!
Entropy \scriptstyle \begin{align}
                      \scriptstyle k &\scriptstyle \,+\, \ln\theta \,+\, \ln[\Gamma(k)]\\
                      \scriptstyle   &\scriptstyle \,+\, (1 \,-\, k)\psi(k)
                    \end{align} \scriptstyle \begin{align}
                      \scriptstyle \alpha &\scriptstyle \,-\, \ln \beta \,+\, \ln[\Gamma(\alpha)]\\
                      \scriptstyle   &\scriptstyle \,+\, (1 \,-\, \alpha)\psi(\alpha)
                    \end{align}
Moment-generating function (mgf) \scriptstyle (1 \,-\, \theta t)^{-k} \text{ for } t \;<\; \frac{1}{\theta}\,\! \scriptstyle \left(1 \,-\, \frac{t}{\beta}\right)^{-\alpha} \text{ for } t \;<\; \beta\,\!
Characteristic function \scriptstyle (1 \,-\, \theta i\,t)^{-k}\,\! \scriptstyle \left(1 \,-\, \frac{i\,t}{\beta}\right)^{-\alpha}\,\!

In probability theory and statistics, the gamma distribution is a two-parameter family of continuous probability distributions. There are two different parameterizations in common use:

  1. With a shape parameter k and a scale parameter θ.
  2. With a shape parameter α = k and an inverse scale parameter β = 1θ, called a rate parameter.

The parameterization with k and θ appears to be more common in econometrics and certain other applied fields, where e.g. the gamma distribution is frequently used to model waiting times. For instance, in life testing, the waiting time until death is a random variable that is frequently modeled with a gamma distribution.[1]

The parameterization with α and β is more common in Bayesian statistics, where the gamma distribution is used as a conjugate prior distribution for various types of inverse scale (aka rate) parameters, such as the λ of an exponential distribution or a Poisson distribution — or for that matter, the β of the gamma distribution itself. (The closely related inverse gamma distribution is used as a conjugate prior for scale parameters, such as the variance of a normal distribution.)

If k is an integer, then the distribution represents an Erlang distribution; i.e., the sum of k independent exponentially distributed random variables, each of which has a mean of θ (which is equivalent to a rate parameter of 1/θ). Equivalently, if α is an integer, then the distribution again represents an Erlang distribution, i.e. the sum of α independent exponentially distributed random variables, each of which has a mean of 1/β (which is equivalent to a rate parameter of β).

The gamma distribution is the maximum entropy probability distribution for a random variable X for which \scriptstyle E(X) \;=\; k\theta \;=\; \alpha/\beta is fixed and greater than zero, and \scriptstyle E[\ln(X)] \;=\; \psi(k) \,+\, \ln(\theta) \;=\; \psi(\alpha) \,-\, \ln(\beta) is fixed (\scriptstyle \psi is the digamma function).[2]

Contents

  Characterization using shape k and scale θ

A random variable X that is gamma-distributed with shape k and scale θ is denoted

X \sim \Gamma(k, \theta) \equiv \textrm{Gamma}(k,\theta )

  Probability density function

  Illustration of the Gamma PDF for parameter values over k and x with θ set to 1, 2, 3, 4, 5 and 6. One can see each θ layer by itself here [1] as well as by k [2] and x. [3].

The probability density function of the gamma distribution can be expressed in terms of the gamma function parameterized in terms of a shape parameter k and scale parameter θ. Both k and θ will be positive values.

The equation defining the probability density function of a gamma-distributed random variable x is


\begin{align}
f(x;k,\theta) &=  \frac{1}{\theta^k}\frac{1}{\Gamma(k)}x^{k-1}e^{-\frac{x}{\theta}} \\
& \text{ for } x \geq 0 \text{ and } k, \theta > 0
\end{align}

(This parameterization is used in the infobox and the plots.)

  Cumulative distribution function

The cumulative distribution function is the regularized gamma function::

 F(x;k,\theta) = \int_0^x f(u;k,\theta)\,du
                      = \frac{\gamma\left(k, \frac{x}{\theta}\right)}{\Gamma(k)} \,

where \scriptstyle \gamma(k,\, x/\theta) is the lower incomplete gamma function.

It can also be expressed as follows, if k is a positive integer (i.e., the distribution is an Erlang distribution):[3]

F(x;k,\theta) = 1-\sum_{i=0}^{k-1} \frac{1}{i!} \left(\frac{x}{\theta}\right)^i e^{-\frac{x}{\theta}} = \sum_{i=k}^{\infty} \frac{1}{i!} \left(\frac{x}{\theta}\right)^i e^{-\frac{x}{\theta}}

  Characterization using shape α and rate β

Alternatively, the gamma distribution can be parameterized in terms of a shape parameter α = k and an inverse scale parameter β = 1θ, called a rate parameter:


\begin{align}
g(x;\alpha,\beta) &= \beta^{\alpha}\frac{1}{\Gamma(\alpha)} x^{\alpha-1} e^{-\beta x} \\
 & \text{ for } x \geq 0 \text{ and } \alpha, \beta > 0
\end{align}

If α is a positive integer, then

 \Gamma(\alpha) = (\alpha - 1)!\,

A random variable X that is gamma-distributed with shape α and scale β is denoted

X \sim \Gamma(\alpha, \beta) \equiv \textrm{Gamma}(\alpha,\beta)

Both parametrizations are common because either can be more convenient depending on the situation.

  Cumulative distribution function

The cumulative distribution function is the regularized gamma function:

 F(x;\alpha,\beta) = \int_0^x f(u;\alpha,\beta)\,du
                      = \frac{\gamma(\alpha, \beta x)}{\Gamma(\alpha)} \,

where \scriptstyle \gamma(\alpha,\, \beta x) is the lower incomplete gamma function.

It can also be expressed as follows, if α is a positive integer (i.e., the distribution is an Erlang distribution):[4]

F(x;\alpha,\beta) = 1-\sum_{i=0}^{\alpha-1} \frac{1}{i!} (\beta x)^i e^{-\beta x} = \sum_{i=\alpha}^{\infty} \frac{1}{i!} (\beta x)^i e^{-\beta x}

  Properties

  Skewness

The skewness depends only on the first parameter ( α ). It approaches a normal distribution when α is large (approximately when α > 10).

  Median calculation

Unlike the mode and the mean which have readily calculable formulas based on the parameters, the median does not have an easy closed form equation. The median for this distribution is defined as the constant x0 such that

{ 1 \over \Gamma( k ) \theta^k } \int_0^{ x_0 } x^{ k - 1 } e^{ - \frac{ x }{ \theta } } dx = { 1 \over 2 }

The ease of this calculation is dependent on the k parameter. This is best achieved by a computer since the calculations can quickly grow out of control.

For the Γ( n + 1, 1 ) distribution the median ( ν ) is known[5] to lie between

 n + \frac{ 2 }{ 3 } < \nu < n + log_e(2)

This estimate has been improved[6]

 \nu = \frac{ 2 }{ 3 } + \frac{ 8 } { 405 n } - \frac{ 64 } { 5103 n^2 } + \frac{ 2^7.23 } { 3^9.25 n^3 } + O(\frac{ 1 }{ n^4 } )


A method of estimating the median for any Gamma distribution has been derived based on the ratio μ /( μ - ν ) which to a very good approximation when α ≥ 1 is a linear function of α.[7] The median estimated by this method is approximately

 \nu \sim \mu \frac{ 3 \alpha - 0.8 }{ 3 \alpha + 0.2 }

where μ is the mean.

  Summation

If Xi has a Γ(ki, θ) distribution for i = 1, 2, ..., N (i.e., all distributions have the same scale parameter θ), then


  \sum_{i=1}^N X_i \sim
    \mathrm{Gamma}  \left( \sum_{i=1}^N k_i, \theta \right) \,\!

provided all Xi' are independent.

The gamma distribution exhibits infinite divisibility.

  Scaling

If

X\,\sim\,\mathrm{Gamma}(k, \theta) \,

then for any c > 0,

cX \sim \mathrm{Gamma}( k, c\theta) \,

Hence the use of the term "scale parameter" to describe θ.


Equivalently, if

X\,\sim\,\mathrm{Gamma}(\alpha, \beta) \,

then for any c > 0,

cX \sim \mathrm{Gamma}( \alpha, \beta/c) \,

Hence the use of the term "inverse scale parameter" to describe β.

  Exponential family

The Gamma distribution is a two-parameter exponential family with natural parameters k − 1 and −1θ (equivalently, α − 1 and −β), and natural statistics X and ln(X).[citation needed]

If the shape parameter α is held fixed, the resulting one-parameter family of distributions is a natural exponential family.[citation needed]

  Logarithmic expectation

One can show that

\mathbb{E}[\ln(X)] = \psi(\alpha) - \ln(\beta)

or equivalently,

\mathbb{E}[\ln(X)] = \psi(k) + \ln(\theta)

where ψ(α) or ψ(k) is the digamma function.


This can be derived[citation needed] using the exponential family formula for the moment generating function of the sufficient statistic, because one of the sufficient statistics of the gamma distribution is \ln x .[citation needed]

  Information entropy

The information entropy can be derived as[citation needed]


\begin{align}
H(x) = \operatorname{E}(-\ln p(X)) &= \operatorname{E}(-\alpha\ln\beta + \ln\Gamma(\alpha) - (\alpha-1)\ln X + \beta X) \\
&= -\alpha\ln\beta + \ln\Gamma(\alpha) - (\alpha-1)\operatorname{E}(\ln X) + \beta \operatorname{E}(X) \\
&= -\alpha\ln\beta + \ln\Gamma(\alpha) - (\alpha-1)(\psi(\alpha) - \ln(\beta)) + \beta \frac{\alpha}{\beta} \\
&= -\alpha\ln\beta + \ln\Gamma(\alpha) - (\alpha-1)\psi(\alpha) +\alpha\ln\beta - \ln\beta + \alpha \\
&= \ln\Gamma(\alpha) - (\alpha-1)\psi(\alpha) - \ln\beta + \alpha \\
&= \alpha - \ln\beta + \ln\Gamma(\alpha) + (1-\alpha)\psi(\alpha)
\end{align}

In the k,θ parameterization, the information entropy is given by[citation needed]

k + \ln\theta + \ln\Gamma(k) + (1-k)\psi(k) .

  Kullback–Leibler divergence

  Illustration of the Kullback–Leibler (KL) divergence for two Gamma PDFs. Here β = β0 + 1 which are set to 1, 2, 3, 4, 5 and 6. The typical asymmetry for the KL divergence is clearly visible.

The Kullback–Leibler divergence (KL-divergence), as with the information entropy and various other theoretical properties, are more commonly[citation needed] seen using the α,β parameterization because of their uses in Bayesian and other theoretical statistics frameworks.

The KL-divergence of \rm{Gamma}(\alpha_p, \beta_p) ("true" distribution) from \rm{Gamma}(\alpha_q, \beta_q) ("approximating" distribution) is given by[8]


\begin{align}
  D_{\mathrm{KL}}(\alpha_p,&\,\beta_p; \alpha_q, \beta_q) = \\
& (\alpha_p-\alpha_q)\psi(\alpha_p) - \log\Gamma(\alpha_p) + \log\Gamma(\alpha_q) \\
& + \alpha_q(\log \beta_p - \log \beta_q) + \alpha_p\frac{\beta_q-\beta_p}{\beta_p}
\end{align}


Written using the k,θ parameterization, the KL-divergence of \rm{Gamma}(k_p, \theta_p) from \rm{Gamma}(k_q, \theta_q) is given by[citation needed]


  D_{\mathrm{KL}}(k_p,\theta_p; k_q, \theta_q) = 
(k_p-k_q)\psi(k_p) - \log\Gamma(k_p) + \log\Gamma(k_q) + k_q(\log \theta_q - \log \theta_p) + k_p\frac{\theta_p - \theta_q}{\theta_q}

  Laplace transform

The Laplace transform of the gamma PDF is

F(s) = \left(1 + \theta s\right)^{-k} = \frac{\beta^\alpha}{(s + \beta)^\alpha} .

  Parameter estimation

  Maximum likelihood estimation

The likelihood function for N iid observations (x1, ..., xN) is

L(k, \theta) = \prod_{i=1}^N f(x_i;k,\theta)\,\!

from which we calculate the log-likelihood function

\ell(k, \theta) = (k - 1) \sum_{i=1}^N \ln{(x_i)} - \sum_{i=1}^N \frac{x_i}{\theta} - Nk\ln{(\theta)} - N\ln{\Gamma(k)}

Finding the maximum with respect to θ by taking the derivative and setting it equal to zero yields the maximum likelihood estimator of the θ parameter:

\hat{\theta} = \frac{1}{kN}\sum_{i=1}^N x_i \,\!

Substituting this into the log-likelihood function gives

\ell = (k-1)\sum_{i=1}^N\ln{(x_i)} - Nk - Nk\ln{\left(\frac{\sum x_i}{kN}\right)} - N\ln[\Gamma(k)] \,\!

Finding the maximum with respect to k by taking the derivative and setting it equal to zero yields

\ln{(k)} - \psi(k) = \ln{\left(\frac{1}{N}\sum_{i=1}^N x_i\right)} - \frac{1}{N}\sum_{i=1}^N\ln{(x_i)} \,\!

where

\psi(k) = \frac{\Gamma'(k)}{\Gamma(k)} \!

is the digamma function.

There is no closed-form solution for k. The function is numerically very well behaved, so if a numerical solution is desired, it can be found using, for example, Newton's method. An initial value of k can be found either using the method of moments, or using the approximation

\ln(k) - \psi(k) \approx \frac{1}{2k}\left(1 + \frac{1}{6k + 1}\right) \,\!

If we let

s = \ln{\left(\frac{1}{N}\sum_{i=1}^N x_i\right)} - \frac{1}{N}\sum_{i=1}^N\ln{(x_i)}\,\!

then k is approximately

k \approx \frac{3 - s + \sqrt{(s - 3)^2 + 24s}}{12s}

which is within 1.5% of the correct value.[citation needed] An explicit form for the Newton-Raphson update of this initial guess is given by Choi and Wette (1969) as the following expression:

k \leftarrow k - \frac{ \ln k - \psi\left(k\right) - s }{ \frac{1}{k} - \psi\;'\left(k\right) }

where \psi' denotes the trigamma function (the derivative of the digamma function).

The digamma and trigamma functions can be difficult to calculate with high precision. However, approximations known to be good to several significant figures can be computed using the following approximation formulae:


\psi\left(k\right) = \begin{cases}
  \ln(k) - \left( 1 + \left[ 1 - \left( 1/10 - 1 / 21k^2 \right) / k^2 \right] / 6k \right) / 2k , \quad k \geq 8 \\
  \psi\left( k + 1 \right) - 1/k, \quad k < 8
\end{cases}

and


\psi\;'\left(k\right) = \begin{cases}
  ( 1 + [ 1 + ( 1 - [ 1/5 - 1 / 7k^2 ] / k^2 ) / 3k ] / 2k ) / k, \quad k \geq 8, \\
  \psi\;'\left( k + 1 \right) + 1/k^2, \quad k < 8
\end{cases}

For details, see Choi and Wette (1969).

  Bayesian minimum mean-squared error

With known k and unknown  \theta , the posterior PDF for theta (using the standard scale-invariant prior for \theta) is

P(\theta | k, x_1, \dots, x_N) \propto 1/\theta \prod_{i=1}^N f(x_i; k, \theta)\,\!

Denoting

 y \equiv \sum_{i=1}^Nx_i , \qquad P(\theta | k, x_1, \dots, x_N) = C(x_i) \theta^{-N k-1} e^{-\frac{y}{\theta}}\!

Integration over θ can be carried out using a change of variables, revealing that 1θ is gamma-distributed with parameters \scriptstyle \alpha \;=\; Nk,\; \beta \;=\; y.

\int_0^{\infty} \theta^{-Nk - 1 + m} e^{-\frac{y}{\theta}}\, d\theta = \int_0^{\infty} x^{Nk - 1 - m} e^{-xy} \, dx = y^{-(Nk - m)} \Gamma(Nk - m) \!

The moments can be computed by taking the ratio (m by m = 0)

E(x^m) = \frac {\Gamma (Nk - m)} {\Gamma(Nk)} y^m \!

which shows that the mean ± standard deviation estimate of the posterior distribution for theta is

 \frac {y} {Nk - 1} \pm \frac {y^2} {(Nk - 1)^2 (Nk - 2)}

  Generating gamma-distributed random variables

Given the scaling property above, it is enough to generate gamma variables with \scriptstyle \theta \;=\; 1 as we can later convert to any value of \scriptstyle \beta with simple division.

Using the fact that a \scriptstyle \Gamma(1,\, 1) distribution is the same as an \scriptstyle Exp(1) distribution, and noting the method of generating exponential variables, we conclude that if \scriptstyle U is uniformly distributed on \scriptstyle (0,\, 1], then −\ln(U) is distributed \scriptstyle \Gamma(1,\, 1) Now, using the "α-addition" property of gamma distribution, we expand this result:

\sum_{k=1}^n {-\ln U_k} \sim \Gamma(n, 1)

where \scriptstyle U_k are all uniformly distributed on \scriptstyle (0,\, 1] and independent. All that is left now is to generate a variable distributed as \scriptstyle \Gamma(\delta,\, 1) for \scriptstyle 0 \;<\; \delta \;<\; 1 and apply the "α-addition" property once more. This is the most difficult part.

Random generation of gamma variates is discussed in detail by Devroye,[9] noting that none are uniformly fast for all shape parameters. For small values of the shape parameter, the algorithms are often not valid.[10] For arbitrary values of the shape parameter, one can apply the Ahrens and Dieter[11] modified acceptance-rejection method Algorithm GD (shape k ≥ 1), or transformation method[12] when 0 < k < 1. Also see Cheng and Feast Algorithm GKM 3[13] or Marsaglia's squeeze method.[14]

The following is a version of the Ahrens-Dieter acceptance-rejection method:[11]

  1. Let \scriptstyle m be 1.
  2. Generate \scriptstyle V_{3m - 2}, \scriptstyle V_{3m - 1} and \scriptstyle V_{3m} as independent uniformly distributed on \scriptstyle (0,\, 1] variables.
  3. If \scriptstyle V_{3m - 2} \;\le\; v_0, where \scriptstyle v_0 \;=\; \frac e {e \;+\; \delta}, then go to step 4, else go to step 5.
  4. Let \scriptstyle \xi_m \;=\; V_{3m - 1}^{1 / \delta}, \ \eta_m \;=\; V_{3m} \xi _m^ {\delta - 1}. Go to step 6.
  5. Let \scriptstyle \xi_m \;=\; 1 \,-\, \ln {V_{3m - 1}}, \ \eta_m \;=\; V_{3m} e^{-\xi_m}.
  6. If \scriptstyle \eta_m \;>\; \xi_m^{\delta - 1} e^{-\xi_m}, then increment \scriptstyle m and go to step 2.
  7. Assume \scriptstyle \xi \;=\; \xi_m to be the realization of \scriptstyle \Gamma (\delta,\, 1).

A summary of this is

 \theta \left( \xi - \sum _{i=1} ^{\lfloor{k}\rfloor} {\ln U_i} \right) \sim \Gamma (k, \theta)

where

  • \scriptstyle \lfloor{k}\rfloor is the integral part of \scriptstyle k,
  • \scriptstyle \xi has been generated using the algorithm above with \scriptstyle \delta \;=\; \{k\} (the fractional part of \scriptstyle k),
  • \scriptstyle U_k and \scriptstyle V_l are distributed as explained above and are all independent.

  Related distributions

  Special cases

  Conjugate prior

In Bayesian inference, the gamma distribution is the conjugate prior to many likelihood distributions: the Poisson, exponential, normal (with known mean), Pareto, gamma with known shape σ, inverse gamma with known shape parameter, and Gompertz with known scale parameter.

The Gamma distribution's conjugate prior is:[15]

p(k,\theta | p, q, r, s) = \frac{1}{Z} \frac{p^{k-1} e^{-\theta^{-1} q}}{\Gamma(k)^r \theta^{k s}}

Where Z is the normalizing constant, which has no closed form solution. The posterior distribution can be found by updating the parameters as follows.

\begin{align}
  p' &= p\prod_i x_i\\
  q' &= q + \sum_i x_i\\
  r' &= r + n\\
  s' &= s + n
\end{align}

Where \scriptstyle n is the number of observations, and \scriptstyle x_i is the \scriptstyle i^{th} observation.

  Compound gamma

If the shape parameter of the gamma distribution is known, but the inverse-scale parameter is unknown, then a gamma distribution for the inverse-scale forms a conjugate prior. The compound distribution, which results from integrating out the inverse-scale has a closed form solution, known as the compound gamma distribution.[16]

  Others

  Applications

The gamma distribution has been used to model the size of insurance claims[citation needed] and rainfalls.[17] This means that aggregate insurance claims and the amount of rainfall accumulated in a reservoir are modelled by a gamma process. The gamma distribution is also used to model errors in multi-level Poisson regression models, because the combination of the Poisson distribution and a gamma distribution is a negative binomial distribution.

In neuroscience, the gamma distribution is often used to describe the distribution of inter-spike intervals.[18] Although in practice the gamma distribution often provides a good fit, there is no underlying biophysical motivation for using it.

In bacterial gene expression, the copy number of a constitutively expressed protein often follows the gamma distribution, where the scale and shape parameter are, respectively, the mean number of bursts per cell cycle and the mean number of protein molecules produced by a single mRNA during its lifetime.[19]

The gamma distribution is widely used as a conjugate prior in Bayesian statistics. It is the conjugate prior for the precision (i.e. inverse of the variance) of a normal distribution. It is also the conjugate prior for the exponential distribution.

  Notes

  1. ^ See Hogg and Craig (1978, Remark 3.3.1) for an explicit motivation
  2. ^ Park, Sung Y.; Bera, Anil K. (2009). "Maximum entropy autoregressive conditional heteroskedasticity model". Journal of Econometrics (Elsevier): 219–230. http://www.wise.xmu.edu.cn/Master/Download/..%5C..%5CUploadFiles%5Cpaper-masterdownload%5C2009519932327055475115776.pdf. Retrieved 2011-06-02. 
  3. ^ Papoulis, Pillai, Probability, Random Variables, and Stochastic Processes, Fourth Edition
  4. ^ Papoulis, Pillai, Probability, Random Variables, and Stochastic Processes, Fourth Edition
  5. ^ Chen J, Rubin H (1986) Bounds for the difference between median and mean of Gamma and Poisson distributions. Statist Probab Lett 4: 281–283
  6. ^ Choi KP (1994) On the medians of Gamma distributions and an equation of Ramanujan. Proc Amer Math Soc 121 (1) 245–251
  7. ^ Banneheka BMSG, Ekanayake GEMUPD (2009) A new point estimator for the median of Gamma distribution. Viyodaya J Science 14:95-103
  8. ^ W.D. Penny, KL-Divergences of Normal, Gamma, Dirichlet, and Wishart densities[full citation needed]
  9. ^ Luc Devroye (1986). Non-Uniform Random Variate Generation. New York: Springer-Verlag. http://luc.devroye.org/rnbookindex.html.  See Chapter 9, Section 3, pages 401–428.
  10. ^ Devroye (1986), p. 406.
  11. ^ a b Ahrens, J. H. and Dieter, U. (1982). Generating gamma variates by a modified rejection technique. Communications of the ACM, 25, 47–54. Algorithm GD, p. 53.
  12. ^ Ahrens, J. H. and Dieter, U. (1974). Computer methods for sampling from gamma, beta, Poisson and binomial distributions. Computing, 12, 223–246. PDF
  13. ^ Cheng, R.C.H., and Feast, G.M. Some simple gamma variate generators. Appl. Stat. 28 (1979), 290-295.
  14. ^ Marsaglia, G. The squeeze method for generating gamma variates. Comput, Math. Appl. 3 (1977), 321-325.
  15. ^ Fink, D. 1995 A Compendium of Conjugate Priors. In progress report: Extension and enhancement of methods for setting data quality objectives. (DOE contract 95‑831).
  16. ^ Dubey, Satya D. (December 1970). "Compound gamma, beta and F distributions". Metrika 16: 27–31. DOI:10.1007/BF02613934. http://www.springerlink.com/content/u750hg4630387205/. 
  17. ^ Aksoy, H. (2000) "Use of Gamma Distribution in Hydrological Analysis", Turk J. Engin Environ Sci, 24, 419 – 428.
  18. ^ J. G. Robson and J. B. Troy, "Nature of the maintained discharge of Q, X, and Y retinal ganglion cells of the cat," J. Opt. Soc. Am. A 4, 2301-2307 (1987)
  19. ^ N. Friedman, L. Cai and X. S. Xie (2006) "Linking stochastic dynamics to population distribution: An analytical framework of gene expression," Phys. Rev. Lett. 97, 168302.

  References

  • R. V. Hogg and A. T. Craig. (1978) Introduction to Mathematical Statistics, 4th edition. New York: Macmillan. (See Section 3.3.)'
  • S. C. Choi and R. Wette. (1969) Maximum Likelihood Estimation of the Parameters of the Gamma Distribution and Their Bias, Technometrics, 11(4) 683–690

  External links

   
               

 

All translations of Gamma distribution


sensagent's content

  • definitions
  • synonyms
  • antonyms
  • encyclopedia

Dictionary and translator for handheld

⇨ New : sensagent is now available on your handheld

   Advertising ▼

sensagent's office

Shortkey or widget. Free.

Windows Shortkey: sensagent. Free.

Vista Widget : sensagent. Free.

Webmaster Solution

Alexandria

A windows (pop-into) of information (full-content of Sensagent) triggered by double-clicking any word on your webpage. Give contextual explanation and translation from your sites !

Try here  or   get the code

SensagentBox

With a SensagentBox, visitors to your site can access reliable information on over 5 million pages provided by Sensagent.com. Choose the design that fits your site.

Business solution

Improve your site content

Add new content to your site from Sensagent by XML.

Crawl products or adds

Get XML access to reach the best products.

Index images and define metadata

Get XML access to fix the meaning of your metadata.


Please, email us to describe your idea.

WordGame

The English word games are:
○   Anagrams
○   Wildcard, crossword
○   Lettris
○   Boggle.

Lettris

Lettris is a curious tetris-clone game where all the bricks have the same square shape but different content. Each square carries a letter. To make squares disappear and save space for other squares you have to assemble English words (left, right, up, down) from the falling squares.

boggle

Boggle gives you 3 minutes to find as many words (3 letters or more) as you can in a grid of 16 letters. You can also try the grid of 16 letters. Letters must be adjacent and longer words score better. See if you can get into the grid Hall of Fame !

English dictionary
Main references

Most English definitions are provided by WordNet .
English thesaurus is mainly derived from The Integral Dictionary (TID).
English Encyclopedia is licensed by Wikipedia (GNU).

Copyrights

The wordgames anagrams, crossword, Lettris and Boggle are provided by Memodata.
The web service Alexandria is granted from Memodata for the Ebay search.
The SensagentBox are offered by sensAgent.

Translation

Change the target language to find translations.
Tips: browse the semantic fields (see From ideas to words) in two languages to learn more.

last searches on the dictionary :

4543 online visitors

computed in 0.046s

   Advertising ▼

I would like to report:
section :
a spelling or a grammatical mistake
an offensive content(racist, pornographic, injurious, etc.)
a copyright violation
an error
a missing statement
other
please precise:

Advertize

Partnership

Company informations

My account

login

registration

   Advertising ▼