Higher Order Derivatives and Taylor Expansions: Example

LECTURE 10
Higher Order Derivatives and Taylor Expansions

1. Higher Order Derivatives
Since a partial derivative of a function f : Rn
makes sense to talk about partial derivatives of partial derivatives; i.e., higher order partial derivatives.
2 f x2 f , f f and f f where f (x, y) = 3x y + x . x x xy x y yx y x
2 2
R is (wherever it exists) again a function from Rn to R it

2 2
Example 10.1. Compute
2f x2
(6xy + 2x) = x = 6y + 2 f x y
f x x
2 f xy
3x2 + 0 = x = 6x f y x
2 f yx
(6xy + 2x) = y = 6x + 0 = 6x
Note that in this example

2 f xy f = yx
2
This is in fact a general phenomenon; the value of a mixed partial derivative does not depend on the order in which the derivatives are taken . Stated more formally;
Theorem 10.2.
then
If
: Rn
is such that all double partial derivatives
2f xi xj
exist and are continous,
2f xi xj
f = x x
j
2. TAYLORS FORMULA FOR FUNCTIONS OF SEVERAL VARIABLES

2. Taylors Formula for Functions of Several Variables
Recall that if f (x) is a function of a single variable that is continuous and dierentiable up to order n + 1 then Taylors theorem says that
f (x) = f (a) + f (a)(x a) +
where the error term 1 2!
f (a)(x a)2 + +
f (n ) (a) n (x a) + Rn (x, a) n!
Rn (x, a)
is given by the formula
Rn (x, a) =
xx a
n!
s f (n+1) (s)ds
and that, moreover, the error term is of order (x a)n+1 . Thus, to order (x a)n we can approximate the function f (x) by the polynomial function
Tn (x) = f (a) + f (a)(x a) +
1 f (a)(x a)2 + + f (n) (a) (x a)n 2! n!
There is an analogous theorem for functions of severa variables. However, since its general statement is a bit messy unless we introduce some new notation, well simply state the rst and second order Taylor formulae
Theorem 10.3.
Let
f : Rn
have continuous partial derivatives up to order
2.
Then we may write
f (x ) = f (a ) +
with the error term
f (a ) (x a ) + R (x , a )
1
R1 (x, a)
going to zero faster that a constant times
as
The rst order Taylor polynomial is the function

T1 (x) = f (a) +
f (a) (x a)
= f (a ) +
f f ( x1 a1 ) + + ( xn a n ) . x1 a xn a
Note that this function is linear in the coordinates of
x.
Its graph is thus a at plane and generalizes the
idea of the best straight line t to a curve : it represents the best at plane approximation to the graph of
f ( x)
near the point
xo .
Theorem
10.4
Let
Rn R have continuous partial derivatives up to order 3.

n i=0
Then we may write
f (x ) = f (a ) +
f ( a ) (x i xi
ai ) + 1 2
i=0 j=0
2f (a) (xi xi xj
a i ) (x j a j ) + R ( x , a )
2
with the error term
R2 (x, a)
going to zero faster that a constant times
as
Example 10.5. Compute the second order Taylor formula for the function f (x, y) = xy + x2 + y2 about the point (1, 1).
We have
f (1, 1) = 1 + 1 + 1 = 3 f y (1,1) f y (1,1)
= (y + 2x + 0)|(1 1) = 3
,
= (x + 0 + 2y)|(1 1) = 3
,
2 f x2 (1,1) 2 f xy (1,1) 2 f y2 (1,1)
= (0 + 2 + 0)|(1 1) = 2
,
2 f yx (1,1)
= (1 + 0 + 0)|(1 1) = 1
, ,
= (0 + 0 + 2)|(1 1) = 2
So 1 +2
+
2 f x2
f (x, y) = f (1, 1) +
f f ( x 1) + (y 1) y (1,1) y (1,1)
(1 ,1)
(x 1)2 +
(1,1)
2f xy
(1 ,1)
(x 1)(y 1)
2 f y2
(1 ,1)
2 f yx
(y 1)(x 1) +
3
(y 1)2
+O
(x, y) (1, 1)
= 3 + 3(x 1) + 3(y 1) + +O
1 2(x 1)2 + 2(x 1)(y 1) + 2(y 1)2 2 (x, y) (1, 1) 3
= 3 + 3(x 1) + 3(y 1) + (x 1)2 + (x 1)(y 1) + (y 1)2 +O (x, y) (1, 1)

3
Below I present another (equivalent) formula for the second order Taylor expansion. Let (x a) be the n-dimensional column vector with components
x a 1 1 x2 a21 (x a ) = ...
xn a n
and let (x a) be the matrix transpose of (x a) (an n-dimensional row vector) (x a)T = (x1 a1 , x2 a2 , , xn an ) . The gradient vector f (a) = Df (a), according to the conventions of Section 2.3 is an n-dimensional row vector;
T
f (a) =
f f (a), f (a), , x ( a) x1 x2 n
Let us now dene the Hessian matrix at the point a as the n n matrix Hf (a) dened by
Hf (a) =
Then we can write 1 f (x) f (a) + f (a) (x a) + (x a)T Hf (a) (x a) + O 2 for the second order Taylor expansion of f about a.
(a ) 2 f x2 x1 (a ) . . . 2 f xn x1 (a )
2f x1 x1
(a) 2 f x2 x2 (a ) .. .
2f x1 x2
2f x1 xn
f xn xn
.. . .. . 2
( a) ( a)
xa

Higher Order Derivatives and Taylor Expansions: Example

Uploaded by

Copyright:

Available Formats

Higher Order Derivatives and Taylor Expansions: Example

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Higher Order Derivatives and Taylor Expansions: Example

Uploaded by

Copyright:

Available Formats

LECTURE 10

Higher Order Derivatives and Taylor Expansions

R is (wherever it exists) again a function from Rn to R it

Example 10.1. Compute

Note that in this example

is such that all double partial derivatives

exist and are continous,

2. TAYLORS FORMULA FOR FUNCTIONS OF SEVERAL VARIABLES

is given by the formula

1 f (a)(x a)2 + + f (n) (a) (x a)n 2! n!

have continuous partial derivatives up to order

Then we may write

going to zero faster that a constant times

The rst order Taylor polynomial is the function

Note that this function is linear in the coordinates of

Its graph is thus a at plane and generalizes the

near the point

Rn R have continuous partial derivatives up to order 3.

Then we may write

with the error term

going to zero faster that a constant times

2. TAYLORS FORMULA FOR FUNCTIONS OF SEVERAL VARIABLES

f (1, 1) = 1 + 1 + 1 = 3 f y (1,1) f y (1,1)

2 f x2 (1,1) 2 f xy (1,1) 2 f y2 (1,1)

1 2(x 1)2 + 2(x 1)(y 1) + 2(y 1)2 2 (x, y) (1, 1) 3

= 3 + 3(x 1) + 3(y 1) + (x 1)2 + (x 1)(y 1) + (y 1)2 +O (x, y) (1, 1)

2. TAYLORS FORMULA FOR FUNCTIONS OF SEVERAL VARIABLES

You might also like