1
$\begingroup$

Theorem 9.21 Suppose $\mathbf{f}$ maps an open set $E\subset R^n$ into $R^m$. Then $\mathbf{f}\in \mathcal{C}^\prime(E)$ if and only if the partial derivatives $D_jf_i$ exist and are continuous on $E$ for $1\leq i\leq m$, $1\leq j\leq n$.

In the proof, it says "for the converse, it suffices to consider the case $m=1$", but I don't know why. Could you help me? Thanks.

$\endgroup$

1 Answer 1

1
$\begingroup$

For this direction, we are showing that a differentiable map $f =(f_1, ..., f_m) \in C^1$ if the partial derivatives exist and are continuous. The reason why it suffices to prove this is the product structure of $\mathbb{R}^m$. A bit more specifically, for any differentiable function $E \to \mathbb{R}^m$ and tangent vector $v$ at $p \in E$, we have $D(f)(v) = (D(f_1)(v),...,D(f_m)(v))$.

The map $D(f)$ varies continuously if and only if each $D(f_i)$ vary continuously (this is just saying a matrix varies continuously if and only if each column varies continuously). It thus suffices that each $D(f_j)$ is continuous, and each $f_j$ is a map from $E \to \mathbb{R}$.

$\endgroup$
3
  • $\begingroup$ Could you give me more details? I cannot understand. $\endgroup$
    – studyhard
    Commented Feb 11, 2023 at 18:34
  • $\begingroup$ Fixed the response. $\endgroup$
    – Alex Nolte
    Commented Feb 11, 2023 at 18:46
  • $\begingroup$ I'm still confused. I can even say the map $D(f)$ varies continuously if and only if each $D_j(f_i)$ vary continuously (this is just saying a matrix varies continuously if and only if each "element" varies continuously). As a result I get the theorem directly and I don't need the proof for m=1. However, Rudin provides a rigorous proof for m=1, so I'm looking for a similar rigorous proof for m=k (using the result of m=1). $\endgroup$
    – studyhard
    Commented Feb 11, 2023 at 19:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .