# Inequality x^2-1 > 0

Printable View

• Mar 23rd 2011, 03:08 PM
bugatti79
Inequality x^2-1 > 0
Folks,

For x^2-1 > 0 solution x>1 and x<-1

For x^2-1< 0 solution -1< x < 1

I can see what the solution is easily but how do we show this mathematically?

For the first one we have x^2 > 1 implies x > + - SQRT 1 but this doesnt make sense...?

Thanks
• Mar 23rd 2011, 03:17 PM
emakarov
x^2 - 1 = (x - 1)(x + 1). A product of two numbers is positive iff the numbers are both negative or both positive, i.e., iff x < 1, x < -1 or x > 1, x > -1. Obviously, x < -1 implies x < 1, and x > 1 implies x > -1, so the product is positive iff x < -1 or x > 1.
• Mar 23rd 2011, 03:40 PM
e^(i*pi)
$x^2-1 = (x-1)(x+1)$

As emakarov says, the product of two expressions are positive if they have the same sign. That is you need to solve over the following intervals

$x > 1\ ,\ -1 \leq x \leq 1 \ ,\ x < -1$

Edit: you can also plot $y=x^2-1$ (it's a simple graph - y=x^2 shifted down 1) and see where it's greater than 0 (the x axis)
• Mar 23rd 2011, 03:43 PM
bugatti79
Quote:

Originally Posted by emakarov
x^2 - 1 = (x - 1)(x + 1). A product of two numbers is positive iff the numbers are both negative or both positive, i.e., iff x < 1, x < -1 or x > 1, x > -1. Obviously, x < -1 implies x < 1, and x > 1 implies x > -1, so the product is positive iff x < -1 or x > 1.

Hi Emakarov

Ok, so my approach is not correct then?

How about 1-ye^x > 0. This implies y e^x <1.

Is this the kind of exercise where you have to infer what x and y should be?

I take it that x and y cannot be both negative real numbers here otherwise we get a positive.
I know y cannot be 0
• Mar 23rd 2011, 03:46 PM
e^(i*pi)
$1 > ye^x$ which gives $y > e^{-x}$ is a special example due to the fact that $e^x > 0$ for all real x

In your original question we need to consider that a negative squared is a positive. For example, if you tried x = -3 in your OP you'd find the inequality is satisfied
• Mar 23rd 2011, 03:58 PM
bugatti79
Quote:

Originally Posted by e^(i*pi)
$x^2-1 = (x-1)(x+1)$

As emakarov says, the product of two expressions are positive if they have the same sign. That is you need to solve over the following intervals

$x > 1\ ,\ -1 \leq x \leq 1 \ ,\ x < -1$

Edit: you can also plot $y=x^2-1$ (it's a simple graph - y=x^2 shifted down 1) and see where it's greater than 0 (the x axis)

Despite its simplicity, I dont understand this. (x-1)(x+1) do not have the same sign, if they were the same sign ie ((x-1)(x-1) it gives x^2 -2x +1..(Doh)
• Mar 23rd 2011, 04:08 PM
emakarov
When we say that x - 1 and x + 1 have the same sign, we mean that both numbers are positive or both are negative. The sign of x - 1 is not necessarily - .

Quote:

Ok, so my approach is not correct then?
Which approach exactly?

Quote:

How about 1-ye^x > 0.
This depends on what you want to do with this expression. This is a property that becomes true or false for each given x and y. What to do with it is a different issue.
• Mar 23rd 2011, 04:44 PM
bugatti79
Quote:

Originally Posted by emakarov
When we say that x - 1 and x + 1 have the same sign, we mean that both numbers are positive or both are negative. The sign of x - 1 is not necessarily - .

Ok, I never heard this before. Thanks

Quote:

Originally Posted by emakarov
Which approach exactly?

If $x^2 > 1 \implies x > \pm \SQRT 1$ ie x can be either > +1 or > -1. That was my initial attempt, how is it wrong?

Quote:

Originally Posted by emakarov
This depends on what you want to do with this expression. This is a property that becomes true or false for each given x and y. What to do with it is a different issue.

Need to determine the range of x and y values for when

$1-y e^x >0$
$1-y e^x <0$
$1-y e^x =0$

SO for the first one, e^(i*pi) qouted y>e^-x. Therefore if x can be any real number then y < 0 for the inequality to hold?

Thats my interpretation...pardon my ignorance!

Thanks
• Mar 23rd 2011, 05:15 PM
e^(i*pi)
Just so I'm not misinterpreted the question is $1-ye^x > 0$.

I added ye^x to both sides: $1 > ye^x \implies ye^x < 1$

The inequality will change sign if we divide by a negative number so normally it's necessary to take care. This is a special example though, because the range of e^x is greater than 0 we can divide without worrying about the direction of the inequality changing.

Dividing by e^x: $y < \dfrac{1}{e^x}$

Finally, due to the laws of exponents $y < e^{-x}$

My apologies I got my inequality wrong before
• Mar 23rd 2011, 07:22 PM
LoblawsLawBlog
Quote:

Originally Posted by bugatti79
If $x^2 > 1 \implies x > \pm \SQRT 1$ ie x can be either > +1 or > -1. That was my initial attempt, how is it wrong?

What you're attempting is valid. You can take the square root of both sides and keep the inequality, but remember that $\sqrt{x^2}=|x|$, so the result should be $|x|>1$.