This preview shows pages 1–2. Sign up to view the full content.
MAS 4105—Nov. 28, 1994
Jacobian Determinants and Multiple Integrals.
The terminology in this handout is the same as in the earlier “Jacobians, Directional
Derivatives, and the Chain Rule” handout. In the current handout, we consider differen
tiable functions
f
:
R
n
→
R
n
(i.e. functions for which the dimensions of the domain and
target spaces are equal.) Thus, for any
a
∈
R
n
, the Jacobian matrix
J
f
(
a
) is a square
matrix, and we can consider its determinant and (potentially) its inverse.
Exercise.
1. Suppose that
f
:
R
n
→
R
n
is differentiable and has a differentiable inverse function
f

1
; i.e. both
f
◦
f

1
and
f

1
◦
f
are the identity map of
R
n
. (Note: do not assume that
f
is linear!) Show that for all
a
∈
R
n
,
J
f

1
(
f
(
a
)) = (
J
f
(
a
))

1
.
(Hint: Chain Rule.)
The discussion I give below does not
prove
anything. It is merely meant as motivation
for why a certain theorem says a certain thing. In order to keep the main ideas from being
lost in details that are better left to Advanced Calculus, I’ll frequently make assumptions
that I won’t attempt to justify here. I’ll also use imprecise language from time to time.
Consider a “box” (parallelepiped)
P
in
R
n
spanned by linearly independent vectors
v
1
,...,
v
n
. Let
P
a
be the translate of this box by the vector
a
:
P
a
=
{
a
+
t
1
v
1
+
...
+
t
n
v
n

0
≤
t
i
≤
1
, i
= 1
,
2
,...,n
}
.
(You will probably want to have a pencil and paper by your side as you read this, to draw
diagrams of all the objects involved.) We want to compare the volume of the image
f
(
P
a
)
to the volume of
P
a
. We saw in class that, for
linear
maps
L
,
L
(
P
) is also a box, and that
vol
n
(
L
(
P
)) =

det
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '09
 RUDYAK

Click to edit the document details