Domain of ln(x+y) | x ≠ 0, y ≠ 0 or (x,y) ≠ (0,0)?

  • Thread starter Thread starter chetzread
  • Start date Start date
  • Tags Tags
    Domain
chetzread
Messages
798
Reaction score
1

Homework Statement


in my notes , it's stated that the domain of ln(x+y) is x+y> 0 ...
Can i write it as x ≠ 0 , y ≠ 0 ? Or should i write it as (x,y) ≠ (0,0) ?(x,y) ≠ (0,0) represent x , and y can be 0 at the same time ?

Whereas x ≠ 0 , y ≠ 0 only represent x can't be equal to 0 ? for example, when x = 0 , y not = 0 , so , ln(x+y) is defined , when y = 0 , x not = 0 , so , ln(x+y) is defined .

Correct me if i am wrong ?

Homework Equations

The Attempt at a Solution

 
Physics news on Phys.org
chetzread said:

Homework Statement


in my notes , it's stated that the domain of ln(x+y) is x+y> 0 ...
Can i write it as x ≠ 0 , y ≠ 0 ? Or should i write it as (x,y) ≠ (0,0) ?(x,y) ≠ (0,0) represent x , and y can be 0 at the same time ?

Whereas x ≠ 0 , y ≠ 0 only represent x can't be equal to 0 ? for example, when x = 0 , y not = 0 , so , ln(x+y) is defined , when y = 0 , x not = 0 , so , ln(x+y) is defined .

Correct me if i am wrong ?

Homework Equations

The Attempt at a Solution

What about the negative numbers?
 
  • Like
Likes chetzread
negative number are ok , as long as x+y >0 , so , Can i write the domain as (x,y) ≠ (0,0) ?
 
Hi Chetzread,
You may be a little confused because typically we see the equation y = f(x).
My advice is to reword the equation to z = f(x + y) = f(u), where u = x + y

So ask yourself what is the domain of f(u)? What values can u take? What values of x and y add to give u?

You are correct, both x and y cannot be equal to 0 at the same time. However, you may be surprised to learn that either x and y can be negative (not both).

Below is a plot of z = ln(x + y) (obtained from Wolfram), where x is the horizontal axis, y is the axis going out of the page, and z is the vertical axis. Note what we have discussed about the cases y<0, x<0, x=0, y=0.

https://www4b.wolframalpha.com/Calculate/MSP/MSP37851g488iahb1eeddb60000639baefd52h542g2?MSPStoreType=image/gif&s=63

EDIT: Sorry I have misunderstood the OP. x, y ≠ (0, 0) seems correct.
 
Last edited by a moderator:
chetzread said:
negative number are ok , as long as x+y >0 , so , Can i write the domain as (x,y) ≠ (0,0) ?
##(x,y) \neq (0,0)## doesn't exclude ##(-1,-1)##. You will have to add ##x+y>0## anyway, or an equivalent notation.
 
  • Like
Likes chetzread and Big Red Car
Big Red Car said:
Hi Chetzread,
You may be a little confused because typically we see the equation y = f(x).
My advice is to reword the equation to z = f(x + y) = f(u), where u = x + y

So ask yourself what is the domain of f(u)? What values can u take? What values of x and y add to give u?

You are correct, both x and y cannot be equal to 0 at the same time. However, you may be surprised to learn that either x and y can be negative (not both).

Below is a plot of z = ln(x + y) (obtained from Wolfram), where x is the horizontal axis, y is the axis going out of the page, and z is the vertical axis. Note what we have discussed about the cases y<0, x<0, x=0, y=0.

https://www4b.wolframalpha.com/Calculate/MSP/MSP37851g488iahb1eeddb60000639baefd52h542g2?MSPStoreType=image/gif&s=63
Yes, I know what x and y can be negative, as long as x+y> 0 , I'm wondering is it wrong to write the domain as write it as (x,y) ≠ (0,0) ?
 
Last edited by a moderator:
Look at Fresh_42's comment, he/she has corrected my edit.

EDIT: You can see it in the plot btw, what Fresh_42 is saying :). Good luck.
 
  • Like
Likes chetzread
fresh_42 said:
##(x,y) \neq (0,0)## doesn't exclude ##(-1,-1)##. You will have to add ##x+y>0## anyway, or an equivalent notation.
so , if i write the limit as ##(x,y) \neq (0,0)## and ##x+y>0## at the same time , then , my ans would be correct ?

How if i write the domain as ##x+y>0## only , without ##(x,y) \neq (0,0)## , is it correct ?
 
chetzread said:
so , if i write the limit as ##(x,y) \neq (0,0)## and ##x+y>0## at the same time , then , my ans would be correct ?

How if i write the domain as ##x+y>0## only , without ##(x,y) \neq (0,0)## , is it correct ?
All you need is x + y > 0. That way, x can be 0, but y can't, and y can be 0, but x can't.

Something you might be missing is that the inequality x + y > 0 is an entire half-plane: all the points in the plane that are above the line y = -x. The origin is not in this set, so you don't need to also specify that ##(x, y) \ne (0, 0)##.
 
  • #10
Mark44 said:
All you need is x + y > 0. That way, x can be 0, but y can't, and y can be 0, but x can't.

Something you might be missing is that the inequality x + y > 0 is an entire half-plane: all the points in the plane that are above the line y = -x. The origin is not in this set, so you don't need to also specify that ##(x, y) \ne (0, 0)##.
so , just leave the final ans as x + y >0 , will do ?
 
  • #11
chetzread said:
so , if i write the limit as ##(x,y) \neq (0,0)## and ##x+y>0## at the same time , then , my ans would be correct ?

How if i write the domain as ##x+y>0## only , without ##(x,y) \neq (0,0)## , is it correct ?
##x+y>0## automatically implies ##(x,y) \neq (0,0)##. It doesn't need to be mentioned. However, ##x=0## or ##y=0## is still possible as long as the other one is positive. They even can be negative as you correctly have said (as long as the other one is positive and larger in its absolute value). You might write ##x > -y## if you like.

What is the purpose of writing the domain other than ##\{(x,y) \in \mathbb{R}^2\; | \; x+y > 0\}##?
 
  • Like
Likes chetzread
  • #12
ok , i have one more question here , for the domain of ln (x^2 + y^2 ) , it it given in my notes that the ans is x ≠ 0 , y ≠ 0

IMO , there 's no need to give x ≠ 0 , y ≠ 0 , so just leave the ans (x^2 + y^2 ) > 0 , will do ?
 
  • #13
chetzread said:
ok , i have one more question here , for the domain of ln (x^2 + y^2 ) , it it given in my notes that the ans is x ≠ 0 , y ≠ 0

IMO , there 's no need to give x ≠ 0 , y ≠ 0 , so just leave the ans (x^2 + y^2 ) > 0 , will do ?
In this case you don't have to deal with negative numbers, because the squares are always positive: ##(-1)^2=1##.
Therefore only ##(x,y)=(0,0)## can violate this condition. So here ##x^2+y^2 > 0## is equivalent to ##(x,y) \neq (0,0)##. As soon as one is unequal zero, the point is with in the domain. Drawn as a picture it is the entire plane without the origin.
 
  • Like
Likes chetzread
  • #14
fresh_42 said:
In this case you don't have to deal with negative numbers, because the squares are always positive: ##(-1)^2=1##.
Therefore only ##(x,y)=(0,0)## can violate this condition. So here ##x^2+y^2 > 0## is equivalent to ##(x,y) \neq (0,0)##. As soon as one is unequal zero, the point is with in the domain. Drawn as a picture it is the entire plane without the origin.
so , just choose anyone of ##x^2+y^2 > 0## or ##(x,y) \neq (0,0)## , the ans would be correct ?
 
  • Like
Likes fresh_42
  • #15
fresh_42 said:
In this case you don't have to deal with negative numbers, because the squares are always positive: ##(-1)^2=1##.
Therefore only ##(x,y)=(0,0)## can violate this condition. So here ##x^2+y^2 > 0## is equivalent to ##(x,y) \neq (0,0)##. As soon as one is unequal zero, the point is with in the domain. Drawn as a picture it is the entire plane without the origin.
how about x ≠ 0 , y ≠ 0 is it wrong to write the limit as x ≠ 0 , y ≠ 0 ?
 
  • #16
chetzread said:
so , just choose anyone of ##x^2+y^2 > 0## or ##(x,y) \neq (0,0)## , the ans would be correct ?
@fresh_42 correct?
 
  • #17
chetzread said:
how about x ≠ 0 , y ≠ 0 is it wrong to write the limit as x ≠ 0 , y ≠ 0 ?
I'm not sure I understand what you mean by limit. The difficulty with ##x \neq 0\, , \,y \neq 0## is, that it is not clear whether it has to hold for both or just for one of them. Therefore ##(x,y) \neq (0,0)## is clear: not both zero at the same time.
 
  • Like
Likes chetzread
  • #18
  • #19
fresh_42 said:
I'm not sure I understand what you mean by limit. The difficulty with ##x \neq 0\, , \,y \neq 0## is, that it is not clear whether it has to hold for both or just for one of them. Therefore ##(x,y) \neq (0,0)## is clear: not both zero at the same time.
sorry , i mean domain as x ≠ 0 and y ≠ 0 ... Here's the original ans that the author gave ...
 
  • #20
chetzread said:
sorry , i mean domain as x ≠ 0 and y ≠ 0 ... Here's the original ans that the author gave ...
Even this is ambiguous: For ##(x,y)=(1,0)## there is ##[\;(x \neq 0) \text{ and } (y \neq 0) \;] = \text{ false }##, but the point is allowed!
Langauge often isn't very clear, especially when it comes to negations and even more if two of them are combined by and or or.
It definitely should be avoided. In general it is always better to express something positively, if possible, rather than using "not". E.g. it is usually harder to prove, that something cannot occur, than it is to prove that something is true. Think of how long it took to prove Fermat's last theorem: more than 300 years!

So ##\{(x,y)\in \mathbb{R}^2 \,\vert \, x^2 +y^2 > 0\}## is better than something with ##x \neq 0 </whatever> y \neq 0##.
The pairing ##(x,y) \neq (0,0)## avoids these troubles.
 
  • #21
fresh_42 said:
Even this is ambiguous: For ##(x,y)=(1,0)## there is ##[\;(x \neq 0) \text{ and } (y \neq 0) \;] = \text{ false }##, but the point is allowed!
Langauge often isn't very clear, especially when it comes to negations and even more if two of them are combined by and or or.
It definitely should be avoided. In general it is always better to express something positively, if possible, rather than using "not". E.g. it is usually harder to prove, that something cannot occur, than it is to prove that something is true. Think of how long it took to prove Fermat's last theorem: more than 300 years!

So ##\{(x,y)\in \mathbb{R}^2 \,\vert \, x^2 +y^2 > 0\}## is better than something with ##x \neq 0 </whatever> y \neq 0##.
The pairing ##(x,y) \neq (0,0)## avoids these troubles.
So , for this type of question , always stick to (x^2 +y^2 ) > 0 will be enough ?
 
  • #22
chetzread said:
So , for this type of question , always stick to (x^2 +y^2 ) > 0 will be enough ?
Yes. It is pretty clear. But so is ##(x,y) \neq (0,0)##. In this case it isn't much of a difference.
However in ##x+y>0## dealing with the cases makes it more complicated.
 
  • #23
fresh_42 said:
However in x+y>0x+y>0x+y>0 dealing with the cases makes it more complicated.
what do you mean ?
 
  • #24
chetzread said:
what do you mean ?
I mean, if you have ##D=\{(x,y) \in \mathbb{R}^2 \,\vert \, x+y > 0\}## and try to describe it like

$$D=\{(x,y) \in \mathbb{R}^2 \,\vert \, \begin{cases} x \neq 0 & \text{ if } y =0 \\ y \neq 0 & \text{ if } x =0
\\ x > -y & \text{ if } y \leq 0 \text{ and } x>0 \\ x > 0 & \text{ if } y \geq 0 \\ y > -x & \text{ if } x \leq 0 \text{ and } y > 0 \\ y > 0 & \text{ if } x \geq 0 \end{cases} \quad \} $$
it is obvious what is clearer. And it took me quite a while to write this. And I'm still not sure, whether there is a mistake in it.
 
  • Like
Likes chetzread
  • #25
chetzread said:
So , for this type of question , always stick to (x^2 +y^2 ) > 0 will be enough ?
It's much clearer to write (x,y) ≠ (0,0) . Especially, anybody grading your solution will be more certain the you understand what is the domain.
 
  • Like
Likes chetzread
Back
Top