If x and y are non-zero reals satisfying x 2 + y 2 = x 2 y 2 , then what is the maximum value of the expression below?
x y 5 x + 1 2 y + 7 x y
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Megh is that x and y are positive reals ?
If So , Then I have another method !
Triangle
Let denote these by standard notation of Triangle ABC
x = a , y = b , x y = c .
So E = c 5 a + 1 2 b + 7 c E = sin C 5 sin A + 1 2 sin B + 7 sin C E = 1 5 sin A + 1 2 sin B + 7 ( ∵ sin C = 1 ) E m a x = 2 0 .
Log in to reply
I took the shameless route and just used calculus. Both your solutions are really nice.
Wow, both approaches are nice!
They are (arguably) equivalent, but I like how they have different motivations. One as "Pythagorean Formula", and another as "Trigonometric Pythagorean Identity"
Log in to reply
I solved this q by algebraic method but still I have a doubt..if we put x=y=0(or h) it satisfies 1st eqn...h being very small no.
If we substitute value h in req. Expression we get 17h+7h^2÷h^2 as infinite..
Log in to reply
@Sachin Arora – Note that x 2 + y 2 = x 2 y 2 is not satisfied in a neighborhood of ( 0 , 0 ) , other than the point itself. In fact, we can show that other than the origin, we must have ∣ x ∣ , ∣ y ∣ ≥ 1 .
I have edited in the condition that x , y = 0 , which is a reasonable assumption given that they appear in the denominator.
In similar situations where there are solutions near ( 0 , 0 ) , the right approach would be to consider the limiting value.
Log in to reply
@Calvin Lin – Thanks.. I want to know that why h(small quantity) is not solution of 1st eqn. Isn't 2h^2=h^4=0 (Sorry, I may sound dumb)
Log in to reply
@Sachin Arora – For h = 0 , we do not have 2 h 2 = h 4 ( = 0 ) .
Making y the subject of the equation, we see that y 2 = x 2 − 1 x 2 . Since the LHS is always positive, this means that the RHS is always positive. In particular, since x 2 > 0 (with x = 0 , we must have x 2 − 1 > 0 . Hence, ∣ x ∣ > 1 .
I loved both your approaches towards the problem, Deepanshu.
Also, I would appreciate Megh for posting such a nice problem.
I think your diagram is mislabeled. If the sin C is 1, C must be the right angle. Further, in standard notation, b is opposite angle B and c is opposite angle C.
Hi, why is the max of 5 sin theta + 12 cos theta, 13?
Log in to reply
It's simple. Apply Cauchy inequality to get
( 5 sin θ + 1 2 cos θ ) 2 ≤ ( 5 2 + 1 2 2 ) ( s i n 2 θ + c o s 2 θ ) .
Forcing the maximum to be 1 3 .
Simple trigonometry. Since 5 and 12 are the legs of a right triangle with hypotenuse 13, we can write 5 = 13 cos phi and 12 = 13 sin phi. Then the expression becomes 13 cos phi sin theta + 13 sin phi cos theta = 13 (cos phi sin theta + sin phi cos theta) = 13 sin (phi + theta). This is clearly maximum when phi + theta = pi/2, and sin = 1.
I got the exact same solution.
if x , y > 0 . by cauchy : 1 3 x y = ( 2 5 + 1 4 4 ) ( x ² + y ² ) ≥ 5 x + 1 2 y so: x y 5 x + 1 2 y + 7 x y = x y 5 x + 1 2 y + 7 ≤ x y 1 3 x y + 7 = 2 0
i too used cauchy :D
I used Cauchy's as well, but I got ( ( 5 ) y 1 + ( 1 2 ) x 1 + ( 1 ) 7 ) 2 ≤ ( 5 2 + 1 2 2 + 1 2 ) ( x 2 1 + y 2 1 + 7 2 )
⇒ ( ( 5 ) y 1 + ( 1 2 ) x 1 + ( 1 ) 7 ) 2 ≤ ( 2 5 + 1 4 4 + 1 ) ( 1 + 4 9 )
⇒ ( ( 5 ) y 1 + ( 1 2 ) x 1 + ( 1 ) 7 ) ≤ 1 0 8 5
What did I do wrong?
Log in to reply
You have merely shown that 1 0 8 5 ≈ 9 2 is an upper bound. For it to be the maximum, you need to find the least upper bound .
One way of showing that it is the least upper bound, is to check the equality conditions. In this case, we will require 5 x = 1 2 y = 7 x y . However, this does not agree with the original condition.
x^2 + y^2 is greater than equal to x^2y^2 and as it is equal to x^2 + y^2 then the solution for the maximum value should be x = y = 2^(0.5)
Q u e s t i o n c a n b e w r i t t e n a s ( y 5 + x 1 2 + 7 ) W e w o u l d o m m i t 7 a s i t s a c o n s t a n t A p p l y i n g c a u c h y ′ s i n e q u a l i t y ( y 5 + x 1 2 ) 2 ≤ ( 2 5 + 1 4 4 ) ( x 2 y 2 x 2 + y 2 ) ( y 5 + x 1 2 ) ≤ 1 3 n o w a d d i n g 7 ∵ ( y 5 + x 1 2 + 7 ) ≤ 2 0
can you please find out for which values of x and y this is true?
A calculus approach: (little longer)
From the equation x 2 + y 2 = x 2 y 2 , we can get that y = x 2 − 1 x . Solving with the other equation we get x 2 − 1 x 2 5 x + 1 2 x 2 − 1 x + 7 x 2 − 1 x 2
This (function) can be simplified to:
x 5 x 2 − 1 + x 1 2 + 7
Knowing that the derivative of the function will be equal to zero at its maximum value, after differentiation and some simplifications we get:
1 2 x 2 − 1 = 5
Then: x = 1 3 1 2 Then, the maximum value of the function will be at x = 1 3 1 2 which is 2 0
Very good and solid approach
This method is general.
This is how I solved it as well. I wish I had seen the sec/csc shortcut but anyway I got the correct answer which was 20. I'm afraid though that David Nasr made a slight transcription error when entering his solution. x is actually equal to 1 2 1 3 not 1 3 1 2 which is clearly incorrect as it would yield an imaginary result in the expression x 2 − 1 .
But if x = 12/13, x^2 - 1 is negative?!?!
Didn't know Cauchy! So, I had to use this method too.
Substituting for sine and cosine would solve the problem. From x 2 + y 2 = X 2 Y 2 , we try to let x 2 y 2 be the square of radius. And x and y are now R c o s ( θ ) and R s i n ( θ ) respectively. Substituting, we obtain ( 5 R c o s ( θ ) + 1 2 R s i n ( θ ) + 7 R ) / R , yielding 5 c o s ( θ ) + 1 2 s i n ( θ ) + 7 . From Cauchy, we get the maximum of 5 c o s ( θ ) + 1 2 s i n ( θ ) , which is square root of [ ( 5 2 + 1 2 2 ) ( c o s 2 ( θ ) + s i n 2 ( θ ) ) ] . Summing up, we get 1 3 + 7 = 2 0 .
Lagrange Multiplier's Method:
Let f ( x , y ) = x y 5 x + 1 2 y + 7 x y = y 5 + x 1 2 + 7
The partial derivatives:
f x = − x 2 1 2 f y = − y 2 5
Let g ( x , y ) = x 2 + y 2 − x 2 y 2
The partial derivatives:
g x = 2 x − 2 x y 2 g y = 2 y − 2 x 2 y
Now, we can equate them:
− x 2 1 2 = λ ( 2 x − 2 x y 2 ) ⟹ Eq. (1)
− y 2 5 = λ ( 2 y − 2 x 2 y ) ⟹ Eq. (2)
Plus the original condition:
x 2 + y 2 = x 2 y 2 ⟹ Eq. (3)
Now, Eq. (1) ÷ Eq. (2):
( − y 2 5 ) ( − x 2 1 2 ) = λ ( 2 y − 2 x 2 y ) λ ( 2 x − 2 x y 2 ) 5 x 2 1 2 y 2 = 2 y ( 1 − x 2 ) 2 x ( 1 − y 2 ) 1 2 y 3 ( 1 − x 2 ) = 5 x 3 ( 1 − y 2 )
From Eq. (3):
x 2 + y 2 = x 2 y 2 ⟹ y 2 = x 2 − 1 x 2
Substitute it in:
1 2 ( x 2 − 1 x 2 ) 3 / 2 ( 1 − x 2 ) = 5 x 3 ( 1 − x 2 − 1 x 2 )
I don't like square roots, so I square the entire equation (even if you don't square it, you will still get the same answers):
1 4 4 ( x 2 − 1 x 2 ) 3 ( 1 − x 2 ) 2 = 2 5 x 6 ( x 2 − 1 − 1 ) 2 ( x 2 − 1 ) 3 1 4 4 x 6 ( x 2 − 1 ) 2 = ( x 2 − 1 ) 2 2 5 x 6 1 4 4 x 6 ( x 2 − 1 ) 4 = 2 5 x 6 ( x 2 − 1 ) 3 1 4 4 x 6 ( x 2 − 1 ) 4 − 2 5 x 6 ( x 2 − 1 ) 3 = 0 x 6 ( x 2 − 1 ) 3 ( 1 4 4 ( x 2 − 1 ) − 2 5 ) = 0 x = 0 , x = ± 1 , x = ± 1 2 1 3
Now, x , y = 0 , so we reject x = 0
For x = ± 1 ⟹ x 2 = 1 , we substitute it into Eq. (3):
1 + y 2 = y 2
There are no solutions for this, so x = ± 1 is not a solution.
For x = ± 1 2 1 3 ⟹ x 2 = 1 4 4 1 6 9 , we substitute it into Eq. (3):
1 4 4 1 6 9 + y 2 = 1 4 4 1 6 9 y 2 1 4 4 2 5 y 2 = 1 4 4 1 6 9 y 2 = 2 5 1 6 9 ⟹ y = ± 5 1 3
We have 4 possible solutions: ( 1 2 1 3 , 5 1 3 ) , ( − 1 2 1 3 , 5 1 3 ) , ( 1 2 1 3 , − 5 1 3 ) , ( − 1 2 1 3 , − 5 1 3 )
Substitute all 4 into f ( x , y ) , and the largest value obtained will be the answer:
f ( 1 2 1 3 , 5 1 3 ) = 2 0 f ( − 1 2 1 3 , 5 1 3 ) ≈ − 2 . 1 5 2 f ( 1 2 1 3 , − 5 1 3 ) ≈ 1 6 . 1 5 4 f ( − 1 2 1 3 , − 5 1 3 ) = − 6
Therefore, the maximum value of the expression is 2 0
We have,
x 2 + y 2 = x 2 y 2
x 2 1 + y 2 1 = 1 ; x = 0 , y = 0
Using the Cauchy-Schwarz inequality,
a 2 + b 2 ≥ l 2 + m 2 l a + m b
Putting a = x 1 , b = y 1 , l = 1 2 , m = 5 , we get
x 2 1 + y 2 1 ≥ 1 3 x 1 2 + y 5
1 ≥ 1 3 x 1 2 + y 5
x 1 2 + y 5 ≤ 1 3
Hence, x y 5 x + 1 2 y + 7 x y = y 5 + x 1 2 + 7 ≤ 1 3 + 7 ≤ 2 0
T h u s , t h e m a x v a l u e o f t h e e x p r e s s i o n i s 2 0 .
-1000000 1 -1 18.999995 -5.000005
2.59 1.084061642 -1.084061642 19.99998317 -2.138979305
2.6 1.083333333 -1.083333333 20 -2.153846154
2.61 1.082614821 -1.082614821 19.99998345 -2.168565821
Excel's answer: 20
The calculus solution can be simplified by making the substitutions a = 1/y, b = 1/x. Then b = sqrt(1-a*a) and E can be expressed in terms of a alone. Differentiation shows that E has a maximum of 20 when a = 5/13.
Can you provide more details? Thanks!
Substitute a = 1/x, b = 1/y. The first equation becomes a^2 + b^2 = 1 and the expression to be maximized becomes E = 5 * b + 12 * a +7 or E = (+/-) 5 * sort(1 - a^2) + 12 * a + 7. Setting the derivative of E to zero and solving for a gives a = (+/-)12/13, b = (+/-)5/13. By inspection the maximum value of E = 20 occurs when a and b are both positive.
Problem Loading...
Note Loading...
Set Loading...
For non geometric solution :
x 2 1 + y 2 1 = 1 ( ∵ x , y = 0 ) .
x = sec θ & y = csc θ .
E = x y 5 x + 1 2 y + 7 x y E = y 5 + x 1 2 + 7 E = 5 sin θ + 1 2 cos θ + 7 E m a x = 1 3 + 7 = 2 0 .
But I believe There must be an geometric Solution ! So I'am working on it when I got correct then I will Surely Post it !
Geometric solution:
Triangle
Let denote these by standard notation of Triangle ABC
x = a , y = b , x y = c .
So E = c 5 a + 1 2 b + 7 c E = sin C 5 sin A + 1 2 sin B + 7 sin C E = 1 5 sin A + 1 2 sin B + 7 ( ∵ sin C = 1 ) E m a x = 2 0 .