MHB Showing that equality of complex numbers implies that they lie on the same ray

Click For Summary
The discussion centers on proving that if the equality of the magnitudes of the sum of complex numbers holds, then the complex numbers must lie on the same ray. The proposed solution uses induction, starting with the base case for two complex numbers, showing that equality occurs when they are collinear. The thread suggests that a more concise proof can be achieved by leveraging properties of complex conjugates and the Cauchy-Schwarz inequality, emphasizing that equality holds only when the complex numbers are aligned. The teacher's feedback indicates that the original argument contained unnecessary details and could be streamlined for clarity. The conclusion is that a simpler approach using the properties of complex numbers can effectively demonstrate the desired result.
kalish1
Messages
79
Reaction score
0
This problem has been on my mind for a while.

----------
**Problem:**

Show that **if**
\begin{equation}
|z_1+z_2+\dots+z_n| = |z_1| + |z_2| + \dots + |z_n|
\end{equation}
**then** $z_k/z_{\ell} \ge 0$ for any integers $k$ and $\ell$, $1 \leq k, \ell \leq n,$ for which $z_{\ell} \ne 0.$

----------
**Proposed solution:**
For the base case $n = 2,$ we want to find a condition for which $|z_1+z_2| = |z_1|+|z_2|.$ From the book and class discussions, we see that equality occurs if $z_1$ and $z_2$ are collinear. Now assume the statement is true for $n=m.$ That is, assume:
\begin{equation}
|z_1+z_2+\dots+z_m| = |z_1| + |z_2| + \dots + |z_m|
\end{equation}
if and only if $z_k/z_{\ell} \ge 0$ for all $1 \le k,\ell \le m$ when $z_{\ell} \neq 0.$

Let $m \ge 2.$ We need to show:
\begin{equation}
|z_1+z_2+\dots+z_m+z_{m+1}| = |z_1| + |z_2| + \dots + |z_m| + |z_{m+1}|
\end{equation}
if and only if $z_k/z_{\ell} \ge 0$ for all $1 \le k,\ell \le m+1$ when $z_{\ell} \neq 0.$

Let $z=z_1+\ldots+z_m.$ We have $$|z_1 + \ldots + z_{m+1}| = |z + z_{m+1}| \overset{(1)}{\leq} |z| + |z_{m+1}| \overset{(2)}{\leq} |z_1| + \ldots + |z_m| + |z_{m+1}|$$

$(1)$: Equality holds if and only if $z/z_{m+1} \ge 0$ when $z_{m+1} \neq 0$ (according to the base case)

$(2)$: Equality holds if and only if $z_k/z_{\ell} \ge 0$ for all $1 \le k,\ell \le m$ when $z_{\ell} \neq 0$ (according to the induction hypothesis)

So what we have so far is:
$$|z_1 + \ldots + z_{m+1}| = |z_1| + \ldots + |z_{m+1}|$$ if and only if

**$\dfrac{z_1 + \ldots + z_m}{z_{m+1}} \ge 0,$ when $z_{m+1} \neq 0,$ and $z_k/z_{\ell} \ge 0$ for all $1 \le k,\ell \le m,$ when $z_{\ell} \neq 0.$**

We need to show $\mathrm{bold \ statement} \implies z_k/z_{\ell} \ge 0 \ \mathrm{for \ all} \ 1 \le k,\ell \le m+1 \ \mathrm{when} \ z_{\ell} \neq 0.$

Assume $\mathrm{bold \ statement}$ holds. We need to show $z_k/z_{\ell} \ge 0$ for $\ell=m+1$ and $1 \le k \le m+1$ (when $z_{m+1} \neq 0$).

When $k=m+1,$ we have $z_{k}/z_{m+1}=z_{m+1}/z_{m+1}=1 \ge 0.$ Assume now that $k \le m.$ Assume that $z_{k} \neq 0.$

From $\mathrm{bold \ statement}$ we know that $z_1/z_k \ge 0, \ldots, z_m/z_k \ge 0$ implies $\dfrac{z_1+ \ldots +z_m}{z_k} \ge 0.$ Note that $$\dfrac{z_1+ \ldots +z_m}{z_k} = 0 \iff z_1 = \dots = z_m = 0.$$ But since $z_k \neq 0,$ we have $\dfrac{z_1+ \ldots +z_m}{z_k} \neq 0.$

Thus
$$\dfrac{z_k}{z_{m+1}} = \underbrace{\dfrac{z_1+\ldots+z_m}{z_{m+1}}}_{\ge 0 \ \mathrm{bold \ statement}}\cdot\dfrac{z_k}{z_1+\ldots+z_m} \ge 0.$$

----------
I wrote this solution, but my teacher did not approve of it, saying that there is too much extraneous information that doesn't answer the problem.

## How can I solve the problem with the following information:
$z_k = t_kz_1,$ for $z_1 \neq 0,$ so that $z_k$ and $z_1$ lie on the same ray for every $k?$

Thanks for any help.

This question has been crossposted here: http://math.stackexchange.com/questions/1200965/showing-that-equality-of-complex-numbers-implies-that-they-lie-on-the-same-ray
 
Physics news on Phys.org
kalish said:
This problem has been on my mind for a while.

----------
**Problem:**

Show that **if**
\begin{equation}
|z_1+z_2+\dots+z_n| = |z_1| + |z_2| + \dots + |z_n|
\end{equation}
**then** $z_k/z_{\ell} \ge 0$ for any integers $k$ and $\ell$, $1 \leq k, \ell \leq n,$ for which $z_{\ell} \ne 0.$

----------
Without reading it through in fine detail, I have the impression that your argument is essentially correct. But it can certainly be considerably shortened. I think that what your teacher has in mind is something like this: $$| z_1 + z_2 + \ldots + z_n|^2 = (z_1 + z_2 + \ldots + z_n)(\bar z_1 + \bar z_2 + \ldots + \bar z_n) = \sum_{j=1}^n|z_j|^2 + \sum_{1\leqslant j<k\leqslant n}2\text{re}(z_j\bar z_k),$$ where the bars denote complex conjugates. But $\text{re}(z_j\bar z_k) \leqslant |z_j||z_k|$, with equality only if $z_j$ and $z_k$ lie on the same ray (that is essentially the result for the base case $n=2$). Therefore $$| z_1 + z_2 + \ldots + z_n|^2 \leqslant \sum_{j=1}^n|z_j|^2 + \sum_{1\leqslant j<k\leqslant n}2 |z_j||z_k| = \bigl(|z_1| + |z_2| + \ldots + |z_n|\bigr)^2,$$ with equality only if all the $z_j$ lie on the same ray. Now all you have to do is to take the square root of both sides, to get the result.
 
We all know the definition of n-dimensional topological manifold uses open sets and homeomorphisms onto the image as open set in ##\mathbb R^n##. It should be possible to reformulate the definition of n-dimensional topological manifold using closed sets on the manifold's topology and on ##\mathbb R^n## ? I'm positive for this. Perhaps the definition of smooth manifold would be problematic, though.

Similar threads

Replies
1
Views
2K
Replies
32
Views
3K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
816
  • · Replies 2 ·
Replies
2
Views
4K
  • · Replies 2 ·
Replies
2
Views
771
  • · Replies 10 ·
Replies
10
Views
3K
Replies
5
Views
2K