Jump to content

Inverse-chi-squared distribution

From Wikipedia, the free encyclopedia
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.
Inverse-chi-squared
Probability density function
Cumulative distribution function
Parameters
Support
PDF
CDF
Mean for
Median
Mode
Variance for
Skewness for
Excess kurtosis for
Entropy

MGF ; does not exist as real valued function
CF

In probability and statistics, the inverse-chi-squared distribution (or inverted-chi-square distribution[1]) is a continuous probability distribution of a positive-valued random variable. It is closely related to the chi-squared distribution. It is used in Bayesian inference as conjugate prior for the variance of the normal distribution.[2]

Definition

The inverse chi-squared distribution (or inverted-chi-square distribution[1] ) is the probability distribution of a random variable whose multiplicative inverse (reciprocal) has a chi-squared distribution.

If follows a chi-squared distribution with degrees of freedom then follows the inverse chi-squared distribution with degrees of freedom.

The probability density function of the inverse chi-squared distribution is given by

In the above and is the degrees of freedom parameter. Further, is the gamma function.

The inverse chi-squared distribution is a special case of the inverse-gamma distribution. with shape parameter and scale parameter .

  • chi-squared: If and , then
  • scaled-inverse chi-squared: If , then
  • Inverse gamma with and
  • Inverse chi-squared distribution is a special case of type 5 Pearson distribution

See also

References

  1. ^ a b Bernardo, J.M.; Smith, A.F.M. (1993) Bayesian Theory, Wiley (pages 119, 431) ISBN 0-471-49464-X
  2. ^ Gelman, Andrew; et al. (2014). "Normal data with a conjugate prior distribution". Bayesian Data Analysis (Third ed.). Boca Raton: CRC Press. pp. 67–68. ISBN 978-1-4398-4095-5.