Scribd
Upload
52 views3 pages

Taylor Expansions in 2d

This document discusses Taylor expansions for functions of two variables. It shows that by treating a function f(x,y) as a function F(t) of a single variable t, with x and y defined in terms of t, the Taylor expansion formulas for a single variable can be applied. Taking derivatives of F(t) and evaluating at t=0 gives terms involving the partial derivatives of f, allowing its expansion around a point (x0,y0). As an example, a second order Taylor expansion of f(x0+Δx, y0+Δy) is given involving its value and first and second partial derivatives at (x0,y0).

Uploaded by

Khairy Elsayed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
52 views3 pages

Taylor Expansions in 2d

This document discusses Taylor expansions for functions of two variables. It shows that by treating a function f(x,y) as a function F(t) of a single variable t, with x and y defined in terms of t, the Taylor expansion formulas for a single variable can be applied. Taking derivatives of F(t) and evaluating at t=0 gives terms involving the partial derivatives of f, allowing its expansion around a point (x0,y0). As an example, a second order Taylor expansion of f(x0+Δx, y0+Δy) is given involving its value and first and second partial derivatives at (x0,y0).

Uploaded by

Khairy Elsayed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Taylor Expansions in 2d

In your first year Calculus course you developed a family


of formulae for approximating a function F (t) for t near any
fixed point t0 .
F (t0 + t) F (t0 )
F (t0 + t) F (t0 ) + F 0 (t0 )t
F (t0 + t) F (t0 ) + F 0 (t0 )t + 12 F 00 (t0 )t2
..
.
2
1 00
F
(t
)t
0
2!
1
+ n!
F (n) (t0 )tn

F (t0 + t) F (t0 ) + F 0 (t0 )t +


+

3
1 (3)
F
(t
)t
0
3!

You may have also found a formula for the error introduced
in making this approximation. The error En (t) is defined
by
2
1 00
F
(t
)t
0
2!
(n)
1
(t0 )tn + En (t)
n! F

F (t0 + t) = F (t0 ) + F 0 (t0 )t +


+ +
and obeys
En (t) =

(n+1)
n+1
1
F
(t
)t
(n+1)!

for some (unknown) t between t0 and t0 + t.


1

We now generalize to functions of more than one variable. Suppose we wish to approximate f (x0 + x, y0 + y)
for x and y near zero. The trick is to write
f (x0 +x, y0 +y) = F (1) with F (t) = f (x0 +tx, y0 +ty)
and think of x0 , y0 , x and y as constants so that F is
a function of the single variable t. Then we can apply our
single variable formulae with t0 = 0 and t = 1. To do so
we need to compute various derivatives of F (t) at t = 0, by
applying the chain rule to

F (t) = f x(t), y(t) with x(t) = x0 + tx, y(t) = y0 + ty
d
dt x(t)

= x and ddt y(t) = y, the chain rule gives


d
d
f
f
dF
(t)
=
x(t),
y(t)
x(t)
+
x(t),
y(t)
dt
x
dt
y
dt y(t)


= fx x(t), y(t) x + fy x(t), y(t) y
h
i
d
d
fx
fx
d2 F
dt2 (t) =
x x(t), y(t) dt x(t) + y x(t), y(t) dt y(t) x
i
h
d
d
fy
fy
+ x x(t), y(t) dt x(t) + y x(t), y(t) dt y(t) y

Since

= fxx x2 + 2fxy xy + fyy y 2

and so on. Its not hard to prove by induction that, in general,


F

(n)

(t) =

n
X

m=0

n n
f
m
nm
m xm y nm (x0 +tx, y0 +ty)x y

n
m

n!
is the standard binomial coefficient.
where
= m!(nm)!
So when t = 0,

F (0) = f (x0 ,y0 )


dF
dt
2
d F
dt2

(0) = fx (x0 ,y0 )x + fy (x0 ,y0 )y


(0) = fxx (x0 ,y0 )x2 + 2fxy (x0 ,y0 )xy + fyy (x0 ,y0 )y 2

Subbing these into



f (x0 + x, y0 + y) = F (t0 + t)
= F (0) +

t0 =0,t=1
F 0 (0) + 12 F 00 (0)

gives

f (x0 + x, y0 + y)
= f (x0 ,y0 )
+ fx (x0 ,y0 )x + fy (x0 ,y0 )y


+ 12 fxx (x0 ,y0 )x2 + 2fxy (x0 ,y0 )xy + fyy (x0 ,y0 )y 2

3
3
+ O x + y

You might also like