Redirected from "Gaussian distribution".
Contents
Context
Measure and probability theory
Contents
Definition
A probability distribution on a Cartesian space is called Gaussian or a normal distribution if it is of the form
where is some matrix such that is a positive definite bilinear form. Here denotes the determinant and is the canonical bilinear form on .
Since is the coordinate of the volume element associated with , we may equivalently write this as
The mean of this distribution is ; for a distribution with mean , replace with .
The matrix is the inverse of the covariance matrix. In particular, for , we may write (or for mean ) in place of , where is the standard deviation; similarly, becomes . This gives the form
In the case that one denotes this distribution by .
Square as a random variable
The square of a random number with probability distribution follows a scaled Chi-squared distribution.
This has positive expected value and bounded variance .
The probability density function of is
with . For , the general moment formulas are:
and
where is the standard normal variable.
Specifically, for , one has and .
If we define the standardized centered variable , then by linearity of expectation:
By scaling properties of variance, dividing by the constant rescales the variance by :
Substituting directly into confirms this independence from :
Because identically for all , the standardized deviation is strictly scale-invariant with zero mean and variance .
References
See also