| Line 21: | Line 21: | ||
</div> | </div> | ||
| − | The univariate normal density is completely specified by two parameters; its mean ''μ '' and variance ''σ<sup>2</sup>''. The function f<sub>x</sub> can be written as ''N(μ,σ)'' which says that ''x'' is distributed normally with mean ''μ'' and variance ''σ<sup>2</sup>''. | + | The univariate normal density is completely specified by two parameters; its mean ''μ '' and variance ''σ<sup>2</sup>''. The function f<sub>x</sub> can be written as ''N(μ,σ)'' which says that ''x'' is distributed normally with mean ''μ'' and variance ''σ<sup>2</sup>''. Samples from normal distributions tend to cluster about the mean with a spread related to the standard deviation ''σ''. |
For the multivariate normal density in ''d'' dimensions, f<sub>x</sub> is written as | For the multivariate normal density in ''d'' dimensions, f<sub>x</sub> is written as | ||
| Line 35: | Line 35: | ||
<div style="margin-left: 25em;"> | <div style="margin-left: 25em;"> | ||
<math>\boldsymbol{\Sigma} = \mathcal{E} \left [(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t \right] = \int(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t p(\mathbf{x})\, dx</math> | <math>\boldsymbol{\Sigma} = \mathcal{E} \left [(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t \right] = \int(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t p(\mathbf{x})\, dx</math> | ||
| + | </div> | ||
| + | |||
| + | where the expected value of a vector or a matrix is found by taking the expected value of the individual components. i.e if ''x<sub>i<\sub>'' is the ''i''th component of '''x''', ''μ<sub>i<\sub>'' the ''i''th component of '''μ''', and ''σ<sub>ij</sub> the ''ij''th component of '''Σ''', then | ||
| + | |||
| + | <div style="margin-left: 25em;"> | ||
| + | <math>\mu_i = \mathcal{E}[x_i] </math> | ||
| + | </div> | ||
| + | |||
| + | and | ||
| + | |||
| + | <div style="margin-left: 25em;"> | ||
| + | <math>\sigma_ij = \mathcal{E}[(x_i - \mu_i)(x_j - \mu_j)] </math> | ||
</div> | </div> | ||
Revision as of 18:05, 4 April 2013
Discriminant Functions For The Normal Density
Lets begin with the continuous univariate normal or Gaussian density.
$ f_x = \frac{1}{\sqrt{2 \pi} \sigma} \exp \left [- \frac{1}{2} \left ( \frac{x - \mu}{\sigma} \right)^2 \right ] $
for which the expected value of x is
$ \mu = \mathcal{E}[x] =\int\limits_{-\infty}^{\infty} xp(x)\, dx $
and where the expected squared deviation or variance is
$ \sigma^2 = \mathcal{E}[(x- \mu)^2] =\int\limits_{-\infty}^{\infty} (x- \mu)^2 p(x)\, dx $
The univariate normal density is completely specified by two parameters; its mean μ and variance σ2. The function fx can be written as N(μ,σ) which says that x is distributed normally with mean μ and variance σ2. Samples from normal distributions tend to cluster about the mean with a spread related to the standard deviation σ.
For the multivariate normal density in d dimensions, fx is written as
$ f_x = \frac{1}{(2 \pi)^ \frac{d}{2} |\boldsymbol{\Sigma}|^\frac{1}{2}} \exp \left [- \frac{1}{2} (\mathbf{x} -\boldsymbol{\mu})^t\boldsymbol{\Sigma}^{-1} (\mathbf{x} -\boldsymbol{\mu}) \right] $
where x is a d-component column vector, μ is the d-component mean vector, Σ is the d-by-d covariance matrix, and |Σ| and Σ-1 are its determinant and inverse respectively. Also, (x -&mu)t denotes the transpose of (x -&mu).
and
$ \boldsymbol{\Sigma} = \mathcal{E} \left [(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t \right] = \int(\mathbf{x} - \boldsymbol{\mu})(\mathbf{x} - \boldsymbol{\mu})^t p(\mathbf{x})\, dx $
where the expected value of a vector or a matrix is found by taking the expected value of the individual components. i.e if xi<\sub> is the ith component of x, μi<\sub> the ith component of μ, and σij the ijth component of Σ, then
$ \mu_i = \mathcal{E}[x_i] $
and
$ \sigma_ij = \mathcal{E}[(x_i - \mu_i)(x_j - \mu_j)] $
