Namespaces
Variants
Actions

Difference between revisions of "Discriminant function"

From Encyclopedia of Mathematics
Jump to: navigation, search
(Importing text file)
 
m (tex encoded by computer)
 
Line 1: Line 1:
A statistic for the construction of a rule for distinguishing in problems of [[Discriminant analysis|discriminant analysis]] involving two distributions. The discrimination problem for two distributions may be stated as follows. Let an observed object with vector of measurements <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332301.png" /> belong to one of the sets <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332302.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332303.png" />, and let it be unknown to which one. The problem is to find a rule which would make it possible to assign the object to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332304.png" /> if the observed value of the vector <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332305.png" /> is known (the discrimination rule). The construction of such a rule is based on subdivision of the sampling space for the vector <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332306.png" /> into domains <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332307.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332308.png" />, such that, if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d0332309.png" /> is found in <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323010.png" />, it is reasonable (from the point of view of the chosen principle for an optimal solution) to assign <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323011.png" /> to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323012.png" />. If the discrimination rule is based on the subdivision <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323013.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323014.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323015.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323016.png" /> are constants and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323017.png" />, the statistic <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323018.png" /> is called the discriminant function, while the domain in which <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/d/d033/d033230/d03323019.png" /> is called the uncertainty zone. Linear discriminant functions, which are simple to handle, are particularly important. If the distributions are normal and have equal covariance matrices, the discriminant function will be linear if the optimality requirements of the discrimination rule are reasonable. The concept of a [[Discriminant informant|discriminant informant]] is introduced in the problem of discriminant analysis with multiple distributions when the [[Bayesian approach|Bayesian approach]] is adopted.
+
<!--
 +
d0332301.png
 +
$#A+1 = 19 n = 0
 +
$#C+1 = 19 : ~/encyclopedia/old_files/data/D033/D.0303230 Discriminant function
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
 +
 
 +
{{TEX|auto}}
 +
{{TEX|done}}
 +
 
 +
A statistic for the construction of a rule for distinguishing in problems of [[Discriminant analysis|discriminant analysis]] involving two distributions. The discrimination problem for two distributions may be stated as follows. Let an observed object with vector of measurements $  x = ( x _ {1} \dots x _ {p} ) $
 +
belong to one of the sets $  \pi _ {i} $,
 +
$  i= 1, 2 $,  
 +
and let it be unknown to which one. The problem is to find a rule which would make it possible to assign the object to $  \pi _ {i} $
 +
if the observed value of the vector $  x $
 +
is known (the discrimination rule). The construction of such a rule is based on subdivision of the sampling space for the vector $  x $
 +
into domains $  R _ {i} $,
 +
$  i= 1, 2 $,  
 +
such that, if $  x $
 +
is found in $  R _ {i} $,  
 +
it is reasonable (from the point of view of the chosen principle for an optimal solution) to assign $  x $
 +
to $  \pi _ {i} $.  
 +
If the discrimination rule is based on the subdivision $  R _ {1} = \{ {x } : {T( x) < a } \} $,  
 +
$  R _ {2} = \{ {x } : {T( x) \geq  b } \} $,  
 +
where $  a $
 +
and $  b $
 +
are constants and $  a < b $,  
 +
the statistic $  T( x) $
 +
is called the discriminant function, while the domain in which $  a \leq  T( x) < b $
 +
is called the uncertainty zone. Linear discriminant functions, which are simple to handle, are particularly important. If the distributions are normal and have equal covariance matrices, the discriminant function will be linear if the optimality requirements of the discrimination rule are reasonable. The concept of a [[Discriminant informant|discriminant informant]] is introduced in the problem of discriminant analysis with multiple distributions when the [[Bayesian approach|Bayesian approach]] is adopted.

Latest revision as of 19:36, 5 June 2020


A statistic for the construction of a rule for distinguishing in problems of discriminant analysis involving two distributions. The discrimination problem for two distributions may be stated as follows. Let an observed object with vector of measurements $ x = ( x _ {1} \dots x _ {p} ) $ belong to one of the sets $ \pi _ {i} $, $ i= 1, 2 $, and let it be unknown to which one. The problem is to find a rule which would make it possible to assign the object to $ \pi _ {i} $ if the observed value of the vector $ x $ is known (the discrimination rule). The construction of such a rule is based on subdivision of the sampling space for the vector $ x $ into domains $ R _ {i} $, $ i= 1, 2 $, such that, if $ x $ is found in $ R _ {i} $, it is reasonable (from the point of view of the chosen principle for an optimal solution) to assign $ x $ to $ \pi _ {i} $. If the discrimination rule is based on the subdivision $ R _ {1} = \{ {x } : {T( x) < a } \} $, $ R _ {2} = \{ {x } : {T( x) \geq b } \} $, where $ a $ and $ b $ are constants and $ a < b $, the statistic $ T( x) $ is called the discriminant function, while the domain in which $ a \leq T( x) < b $ is called the uncertainty zone. Linear discriminant functions, which are simple to handle, are particularly important. If the distributions are normal and have equal covariance matrices, the discriminant function will be linear if the optimality requirements of the discrimination rule are reasonable. The concept of a discriminant informant is introduced in the problem of discriminant analysis with multiple distributions when the Bayesian approach is adopted.

How to Cite This Entry:
Discriminant function. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Discriminant_function&oldid=15604
This article was adapted from an original article by N.M. MitrofanovaA.P. Khusu (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article