The Gamma distribution with shape and rate has probability density functionThe gamma function normalizes its integral to one. Its mean is and its variance is , obtained by integrating powers of . At it is the exponential distribution; integer gives the waiting time to the th event of a rate- Poisson process. Independent gamma variables with the same rate add their shapes. Some sources use scale instead of rate, so the parameter convention must accompany the formula.
In the shape-rate convention for a gamma distribution, the rate parameter appears in the exponential decay:It is the reciprocal of the scale parameter , not the shape . If has this distribution, then has shape and rate one. This follows by the change of variables formula for a probability density.
The mean is and the variance is . Thus doubling the rate at fixed shape halves the mean and quarters the variance. When measures time, has units of inverse time. For positive integer shape , can be the waiting time for the th event of a rate- Poisson process, so the parameter is literally its event rate. Shape-rate and shape-scale software conventions must be distinguished before substituting numerical parameters.
For the rate-parameterized Gamma distribution with shape and rate , integration of its probability density giveswhen , by the gamma function integral and a change of variables. The integral is infinite for . In particular the moment-generating function is finite on an open interval containing zero, so the uniqueness theorem for moment-generating functions applies. Multiplication of these moment-generating functions proves additivity of independent gamma distributions with a common rate.
The probability density function of a Gamma distribution with shape and rate . Integration against gives when ; the integral diverges for . Rate and scale are reciprocal conventions and must be distinguished.
For , the Wilson-Hilferty transformation uses the approximationThe cube root reduces the right skewness of the chi-squared distribution. Inverting gives the approximate -quantile , where is a standard normal quantile. This is an approximation, especially imperfect for extreme tails or small degrees of freedom. For a Gamma distribution with shape and rate , apply it to , which has .[ref]
A gamma distribution translated by a real constant . With shape and rate , its support is and its first three cumulants are , , and . Its skewness is . A negative shift permits negative values; introducing a shift does not automatically respect the nonnegative support of an insurance loss.
If and are independent random variables with Gamma distributions of shapes and the same rate , then has the Beta distribution with parameters , and has the Gamma distribution of shape and rate . Moreover and are independent. The inverse change of variables has absolute Jacobian determinant , which makes the joint probability density factorize on the product domain , . The common rate is essential for this factorization.
If are independent exponential random variables with one common rate , put . The change of variables for probability densities gives on . Its marginal density is , so the conditional probability density of given is on . This is the shape- case of beta-gamma independence. In particular is uniform on and independent of . Equality of the rates is essential: different rates leave an exponential tilt across the conditional interval.
If independent random variables have gamma distributions with shapes and one common rate , thenTheir Laplace transforms of nonnegative random variables multiply to , which is the transform of the displayed gamma distribution; uniqueness of the Laplace transform proves the claim.
Writing , , and puts the gamma density into exponential dispersion family formIts mean is and its variance is .
The inverse-gamma distribution with shape and scale has probability density functionEquivalently, has a gamma distribution with shape and rate . The inverse change of variables has absolute Jacobian determinant , which turns the gamma density into the displayed expression. Thus the inverse-gamma scale parameter is the reciprocal variable's gamma rate, not its gamma scale.
Its moments satisfy when . For positive the defining integral diverges at infinity. In particular,For , the expected value and variance are both one. The distribution is useful for positive scale uncertainty, but its heavy tail means that a valid normalized probability distribution can still lack a finite expected value.
Let be conditionally independent inverse-gamma distribution observations with known shape , unknown scale , and observed values . Their likelihood function isA gamma distribution prior with shape and rate is a conjugate prior, givingThis follows by multiplying the two powers of and the two exponential factors. It uses reciprocal observations, rather than their sample mean.
If , the conditional claim expected value is . Its Bayes estimator under squared error loss is the posterior expected valueHere the target is the conditional expected value, whose posterior second moment is finite because the posterior for is gamma. When , the conditional claim expected value is infinite and this finite-estimate formula does not apply. For example, , , and observations give posterior shape , rate , and estimated conditional claim expected value .
The sum of independent exponential variables of rate has the Erlang distribution with densityConvolving with one more exponential proves the formula inductively.
Let be a positive integer, , and independent uniform random variables on . Thenhas a gamma distribution of shape and rate , also called an Erlang distribution. The unit interval is essential: an unspecified uniform distribution on another interval would not give this algorithm.
For , . Each summand therefore has an exponential distribution with rate , and additivity of independent gamma distributions with a common rate proves the result. For , the output has expected value and variance . The formula directly covers integer shapes; replacing by a noninteger number of draws does not define a sampler for a general gamma shape.
Articles by others on the same topic
The Gamma distribution is a continuous probability distribution defined by two parameters: shape (often denoted as \( k \) or \( \alpha \)) and scale (denoted as \( \theta \) or \( \beta \)). It is widely used in various fields, including statistics, finance, and engineering, due to its ability to model waiting times and processes that are characterized by events that occur independently at a constant average rate.