MedVision ad

inequality (1 Viewer)

xiao1985

Active Member
Joined
Jun 14, 2003
Messages
5,704
Gender
Male
HSC
N/A
PremusDog said:
hey

how do you Show that

x1 + x2 > root of (x1x2)

Thanks!
the inequality does nto stand...

try
x1 = -1
x2 = -1

x1 + x 2 = - 2

sqrt ( -1 x -1) = 1

-2>1 (?) contradiction... therefore the arguemtn is nto true....
 

redruM

Breathe and Stop
Joined
May 11, 2004
Messages
3,954
Gender
Male
HSC
2003
it would stand if there was a condition like x<sub>1</sub> and x<sub2</sub> > 0, or < 0.

i dont know...havent done maths in ages.
 

redruM

Breathe and Stop
Joined
May 11, 2004
Messages
3,954
Gender
Male
HSC
2003
square both sides....and expand LHS

LHS = x<sub>1</sub><sup>2</sup> + 2 x<sub>1</sub>x<sub>2</sub> + x<sub>2</sub><sup>2</sup>

using convention(?)
x<sub>1</sub><sup>2</sup> is always +ve
similarly,
x<sub>2</sub><sup>2</sup> is always +ve

and since x<sub>1</sub> > 1 and x<sub>2</sub> > 1, 2x<sub>1</sub>x<sub>2</sub> is always +ve.

.'. LHS > RHS

i know there will be some flaws in my explanation, but i think my idea is correct.

edit: made some adjusments.
 

xiao1985

Active Member
Joined
Jun 14, 2003
Messages
5,704
Gender
Male
HSC
N/A
ok, in that case =)

edit: >< a min late =p
 
Last edited:

Affinity

Active Member
Joined
Jun 9, 2003
Messages
2,062
Location
Oslo
Gender
Undisclosed
HSC
2003
since when did you start frequenting this forum xiao?
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
Shorter: If x<sub>1</sub> > 1 and x<sub>2</sub> > 1, then (&radic;x<sub>1</sub> - &radic;x<sub>2</sub>)<sup>2</sup> > 0
Hence, x<sub>1</sub> + x<sub>2</sub> > 2&radic;(x<sub>1</sub>x<sub>2</sub>)

Now, 2&radic;(x<sub>1</sub>x<sub>2</sub>) > &radic;(x<sub>1</sub>x<sub>2</sub>) as x<sub>1</sub>x<sub>2</sub> > 0 [you can square root and then add &radic;(x<sub>1</sub>x<sub>2</sub>) to prove this, if needed.]

So, x<sub>1</sub> + x<sub>2</sub> > &radic;(x<sub>1</sub>x<sub>2</sub>), as required.
 

xiao1985

Active Member
Joined
Jun 14, 2003
Messages
5,704
Gender
Male
HSC
N/A
Affinity said:
since when did you start frequenting this forum xiao?
since the great affinity convinced me so... =D

well, to be honest, since the beauty of mathematics truely touched me =p

ok ok, confession: since i was really bored last nite ><
 

CrashOveride

Active Member
Joined
Feb 18, 2004
Messages
1,488
Location
Havana
Gender
Undisclosed
HSC
2006
a b and c are any three +ve real numbers then (a+b+c)/3 >= cube root of(abc)
a+b+c = 1, show 1/a + 1/b + 1/c >= 9
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
CrashOveride said:
a b and c are any three +ve real numbers then (a+b+c)/3 >= cube root of(abc)
One approach:

Let x, y and z be positive reals.

(a) Show that x<sup>2</sup> + y<sup>2</sup> &ge; 2xy

(b) Hence, show that x<sup>2</sup> + y<sup>2</sup> + z<sup>2</sup> - xy - yz - zx &ge; 0

(c) By multiplying the expression in (b) by x + y + z, and making a substitution, show that for a, b and c positive reals:
(a + b + c) / 3 &ge; cubert(abc)​

I suggest you KNOW THIS PROOF. The AM-GM inequailty pops up quite often, and it is very useful to be able to write out proofs for n = 2, 3 and 4.
 

CM_Tutor

Moderator
Moderator
Joined
Mar 11, 2004
Messages
2,642
Gender
Male
HSC
N/A
CrashOveride said:
a+b+c = 1, show 1/a + 1/b + 1/c >= 9
First, show that, for reals x and y, (x / y) + (y / x) = (x<sup>2</sup> + y<sup>2</sup>) / xy &ge; 2

Then, consider the expansion of (a + b + c) * [(1 / a) + (1 / b) + 1 / c)]
 

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top