Namespaces
Variants
Actions

Difference between revisions of "Partial derivative"

From Encyclopedia of Mathematics
Jump to: navigation, search
m (fixing superscripts)
m (fixing dots)
 
(One intermediate revision by the same user not shown)
Line 15: Line 15:
 
The [[Derivative|derivative]] of the function with respect to one of the variables, keeping the remaining variables fixed. For example, if a function  $  f ( x _ {1}, \dots, x _ {n} ) $
 
The [[Derivative|derivative]] of the function with respect to one of the variables, keeping the remaining variables fixed. For example, if a function  $  f ( x _ {1}, \dots, x _ {n} ) $
 
is defined in some neighbourhood of a point  $  ( x _ {1}  ^ {(0)}, \dots, x _ {n}  ^ {(0)} ) $,  
 
is defined in some neighbourhood of a point  $  ( x _ {1}  ^ {(0)}, \dots, x _ {n}  ^ {(0)} ) $,  
then the partial derivative  $  ( \partial  f / \partial  x _ {1} ) ( x _ {1}  ^ {(0)} \dots x _ {n}  ^ {(0)} ) $
+
then the partial derivative  $  ( \partial  f / \partial  x _ {1} ) ( x _ {1}  ^ {(0)}, \dots, x _ {n}  ^ {(0)} ) $
 
of  $  f $
 
of  $  f $
 
with respect to the variable  $  x _ {1} $
 
with respect to the variable  $  x _ {1} $
Line 57: Line 57:
 
$$ \tag{* }
 
$$ \tag{* }
  
\frac{\partial  ^ {m} f }{\partial  x _ {1} ^ {m _ {1} } \dots \partial  x _ {n} ^ {m _ {n} } }
+
\frac{\partial  ^ {m} f }{\partial  x _ {1} ^ {m _ {1} } \cdots \partial  x _ {n} ^ {m _ {n} } }
 
  ,\ \  
 
  ,\ \  
m _ {1} + \dots + m _ {n} = m ,
+
m _ {1} + \cdots + m _ {n} = m ,
 
$$
 
$$
  
Line 67: Line 67:
 
$$  
 
$$  
  
\frac{\partial  ^ {k- 1} f }{\partial  x _ {1} ^ {k _ {1} } \dots \partial  x _ {i} ^ {k _ {i} } \dots
+
\frac{\partial  ^ {k- 1} f }{\partial  x _ {1} ^ {k _ {1} } \cdots \partial  x _ {i} ^ {k _ {i} } \cdots
 
\partial  x _ {n} ^ {k _ {n} } }
 
\partial  x _ {n} ^ {k _ {n} } }
 
  ,\ \  
 
  ,\ \  
k _ {1} + \dots + k _ {n} = k - 1 ,
+
k _ {1} + \cdots + k _ {n} = k - 1 ,
 
$$
 
$$
  
Line 77: Line 77:
 
$$  
 
$$  
  
\frac{\partial  ^ {k} f }{\partial  x _ {1} ^ {k _ {1} } \dots \partial  x _ {i} ^ {k _ {i} + 1 } \dots \partial  x _ {n} ^ {k _ {n} } }
+
\frac{\partial  ^ {k} f }{\partial  x _ {1} ^ {k _ {1} } \cdots \partial  x _ {i} ^ {k _ {i} + 1 } \cdots \partial  x _ {n} ^ {k _ {n} } }
 
  =
 
  =
 
$$
 
$$
Line 87: Line 87:
 
  \left (  
 
  \left (  
 
\frac{\partial  ^ {k- 1} f }{\partial  
 
\frac{\partial  ^ {k- 1} f }{\partial  
x _ {1} ^ {k _ {1} } \dots \partial  x _ {i} ^ {k _ {i} } {} \dots \partial  x _ {n} ^ {k _ {n} } }
+
x _ {1} ^ {k _ {1} } \cdots \partial  x _ {i} ^ {k _ {i} } \cdots \partial  x _ {n} ^ {k _ {n} } }
 
  \right ) .
 
  \right ) .
 
$$
 
$$

Latest revision as of 07:57, 13 May 2022


of the first order of a function in several variables

The derivative of the function with respect to one of the variables, keeping the remaining variables fixed. For example, if a function $ f ( x _ {1}, \dots, x _ {n} ) $ is defined in some neighbourhood of a point $ ( x _ {1} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) $, then the partial derivative $ ( \partial f / \partial x _ {1} ) ( x _ {1} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) $ of $ f $ with respect to the variable $ x _ {1} $ at that point is equal to the ordinary derivative $ ( d f /d x _ {1} ) ( x _ {1} , x _ {2} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) $ at the point $ x _ {1} ^ {( 0)} $ of the function $ f ( x _ {1} , x _ {2} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) $ in the single variable $ x _ {1} $. In other words,

$$ \left . \frac{\partial f }{\partial x _ {1} } ( x _ {1} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) = \frac{d f }{d x _ {1} } ( x _ {1} , x _ {2} ^ {(0)} \dots x _ {n} ^ {(0)} ) \right | _ {x _ {1} = x _ {1} ^ {(0)} } = $$

$$ = \ \lim\limits _ {\Delta x _ {1} \rightarrow 0 } \frac{\Delta _ {x _ {1} } f ( x _ {1} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) }{\Delta x _ {1} } , $$

where

$$ \Delta _ {x _ {1} } f ( x _ {1} ^ {(0)} , \dots, x _ {n} ^ {(0)} ) = $$

$$ = \ f ( x _ {1} ^ {(0)} + \Delta x _ {1} , x _ {2} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) - f ( x _ {1} ^ {(0)}, \dots, x _ {n} ^ {(0)} ) . $$

The partial derivatives

$$ \tag{* } \frac{\partial ^ {m} f }{\partial x _ {1} ^ {m _ {1} } \cdots \partial x _ {n} ^ {m _ {n} } } ,\ \ m _ {1} + \cdots + m _ {n} = m , $$

of order $ m > 1 $ are defined by induction: If the partial derivative

$$ \frac{\partial ^ {k- 1} f }{\partial x _ {1} ^ {k _ {1} } \cdots \partial x _ {i} ^ {k _ {i} } \cdots \partial x _ {n} ^ {k _ {n} } } ,\ \ k _ {1} + \cdots + k _ {n} = k - 1 , $$

has been defined, then by definition

$$ \frac{\partial ^ {k} f }{\partial x _ {1} ^ {k _ {1} } \cdots \partial x _ {i} ^ {k _ {i} + 1 } \cdots \partial x _ {n} ^ {k _ {n} } } = $$

$$ = \ \frac \partial {\partial x _ {i} } \left ( \frac{\partial ^ {k- 1} f }{\partial x _ {1} ^ {k _ {1} } \cdots \partial x _ {i} ^ {k _ {i} } \cdots \partial x _ {n} ^ {k _ {n} } } \right ) . $$

The partial derivative (*) is also denoted by $ D _ {m _ {1} \dots m _ {n} } ^ {m} f $. A partial derivative (*) in which at least two distinct indices $ m _ {i} $ are non-zero is called a mixed partial derivative; otherwise, that is, if the partial derivative has the form $ \partial ^ {m} f / \partial x _ {i} ^ {m} $, it is called unmixed. Under fairly broad conditions, mixed partial derivatives do not depend on the order of differentiation with respect to the different variables. This holds, for example, if all the partial derivatives under consideration are continuous.

If in the definition of a partial derivative the usual notion of a derivative is replaced by that of a generalized derivative in some sense or another, then the definition of a generalized partial derivative is obtained.

Comments

For references see Differential calculus.

How to Cite This Entry:
Partial derivative. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Partial_derivative&oldid=51931
This article was adapted from an original article by L.D. Kudryavtsev (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article