Triangle Inequality: use to prove convergence

Click For Summary
The discussion centers on the application of the triangle inequality to establish bounds for convergence in a mathematical context. Participants clarify the use of the inequalities |x+y| ≤ |x| + |y| and |x-y| ≥ |x| - |y|, focusing on deriving upper and lower bounds for expressions involving z and w. There is confusion regarding how to express bounds in terms of multiples of |w|, particularly when |z| is less than R and |w| is greater than 2R. The conversation emphasizes the need to manipulate these inequalities correctly to find appropriate bounds for the expressions in question. Ultimately, the goal is to ensure that the derived bounds are consistent and useful for proving convergence.
binbagsss
Messages
1,291
Reaction score
12

Homework Statement



Attached

psiconvergence.png


I understand the first bound but not the second.

I am fine with the rest of the derivation that follows after these bounds,

Homework Equations



I have this as the triangle inequality with a '+' sign enabling me to bound from above:

##|x+y| \leq |x|+|y| ## (1)
##|x-y| \geq |x|-|y| ## (2)

and this as the triangle inequality with a '-' sign enabling me to bound from below:

The Attempt at a Solution

So for the first bound we have:

##|z-w| \geq |z| - |w| ##

since we have a strict less than inequality for |z| and a strict greater than equality for |w| , both of these are consistent and we indeed loose the equality option in the triangle inequality to get ##|z-w| > -R ##

I am stuck on the second bound however.
1) I ionly have a upper bound for a subtraction and not a lower via the triangle inequalities. can i get a upper bound from (1) and (2)?

i.e. are you allowewd to do ##|z+(-2w)| \leq |z|+|-2w| ##?

(even if I am, unlike the lower bound, where it turns out the bound we have on ##z## and ##w## are consistent with the triangle inequality, (enabling us to loose the equality and get strictness) there is contrast between the inequalities in this case. ( both items on the right hand side would need lower bounds (or one upper and one equality) but z has a upper bound).

Many thanks .
 

Attachments

  • psiconvergence.png
    psiconvergence.png
    25.3 KB · Views: 1,010
Physics news on Phys.org
binbagsss said:
i.e. are you allowewd to do ##|z+(-2w)| \leq |z|+|-2w| ##?
Yes that is a standard application of the triangle inequality.

(even if I am, unlike the lower bound, where it turns out the bound we have on ##z## and ##w## are consistent with the triangle inequality, (enabling us to loose the equality and get strictness) there is contrast between the inequalities in this case. ( both items on the right hand side would need lower bounds (or one upper and one equality) but z has a upper bound).
Note that the lower bounds are to be expressed as multiples of ##|\omega|##. It is trivial to get such a lower bound for the second term on the RHS. The two inequalities given in the problem enable us to get a lower bound in terms of ##|\omega|## for the first term.
 
andrewkirk said:
Yes that is a standard application of the triangle inequality.
Yes that is a standard application of the triangle inequality.

Ok thanks so from similar manipulations with (2) I see that ##|x+y| ## and ##|x-y|## have the same upper and lower bounds from the triangle inequality which makes sense when I think about what the mod function does and possible combinations etc.

andrewkirk said:
The two inequalities given in the problem enable us to get a lower bound in terms of ##|\omega|## for the first term.

That is fine.

andrewkirk said:
Note that the lower bounds are to be expressed as multiples of ##|\omega|##. It is trivial to get such a lower bound for the second term on the RHS.

I still don't understand this. Why is it trivial?
As I said in my OP there is contrast between the inequalities in this case. ( both items on the right hand side would need lower bounds (or one upper and one equality) but z has a upper bound.
 
binbagsss said:
I still don't understand this. Why is it trivial?
We are looking for upper bounds, not lower bounds.
We want to find an upper bound for ##|z|+|-2w|## that is no greater than ##\frac52|w|##.

The second term is ##|-2w|## and we want to find an upper bound for it that is a multiple of ##|w|##. Given that an equality is also a upper bound, can you express ##|-2w|## as a multiple of ##|w|## and thereby have an upper bound for ##|-2w|##?

For the first term ##|z|##, the two inequalities you've been given are ##|z|<R## and ##|w|>2R##. How can you use those to get an upper bound for ##|z|## that is a multiple of ##|w|##?
 
andrewkirk said:
We are looking for upper bounds, not lower bounds.
We want to find an upper bound for ##|z|+|-2w|## that is no greater than ##\frac52|w|##.

The second term is ##|-2w|## and we want to find an upper bound for it that is a multiple of ##|w|##. Given that an equality is also a upper bound, can you express ##|-2w|## as a multiple of ##|w|## and thereby have an upper bound for ##|-2w|##?

For the first term ##|z|##, the two inequalities you've been given are ##|z|<R## and ##|w|>2R##. How can you use those to get an upper bound for ##|z|## that is a multiple of ##|w|##?

No but the only information is ##|w|>2R##, ##|z|<R## , I can see clearly that the aim is to express in multiplies of ##w## and I'm pretty sure I said I UNDERSTAND the first inequality so unsure why everyone is replying to that, but anyway, given this all we know is ##|w|>2R## , it could be infinity, that's all we know? how do we bound that?
 
binbagsss said:
No but the only information is ##|w|>2R##, ##|z|<R##
Put those two together to get an inequality relation between |z| and |w|. Then use that to get an upper bound on |z|+|-2w| (first converting the second term of that into a multiple of |w|).
 
andrewkirk said:
Put those two together to get an inequality relation between |z| and |w|. Then use that to get an upper bound on |z|+|-2w| (first converting the second term of that into a multiple of |w|).

I don't believe I'd have any issue with going from a bound between |z| and |w| to one of |z|+|-2w| . My issue in my previous still stands whether I am upper bounding - I assume this is what you meant by an inequality - |z|+|-2w| or |z|+|w| - as far as I can see z can be as large as it likes and so I can't see how we can upper bound anything involving an addition of z. Ta.
 
binbagsss said:
I don't believe I'd have any issue with going from a bound between |z| and |w| to one of |z|+|-2w| . My issue in my previous still stands whether I am upper bounding - I assume this is what you meant by an inequality - |z|+|-2w| or |z|+|w| - as far as I can see z can be as large as it likes and so I can't see how we can upper bound anything involving an addition of z. Ta.
?
 
$$|z| < R < 2R < |\omega| \text{ ?}$$
 
  • #10
LCKurtz said:
$$|z| < R < 2R < |\omega| \text{ ?}$$

Apologies typo, \omega can be as large as it wants
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 9 ·
Replies
9
Views
2K
  • · Replies 3 ·
Replies
3
Views
1K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 2 ·
Replies
2
Views
815
  • · Replies 2 ·
Replies
2
Views
2K
Replies
5
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
Replies
9
Views
1K