The sum of 2016 real numbers is 0. If none of the numbers exceeds 1, then what is the maximum possible sum of their cubes?
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.
Let i = 0 ∑ 2 0 1 6 x i = 0 and each x i ≤ 1 .So, there will some positive values and some negative values.
While adding the cubes of the terms then there will be still negative values and positive values.
Assume there are k positive terms and 2 0 1 6 − k negative terms.To maximize positive values we can assign k terms will equal to 1 .So, total value k × 1 = k .So, rest of the values will be negative and should be − k .So, minimum x i for k + 1 ≤ i ≤ 2 0 1 6 is ( 2 0 1 6 − k ) x i = − k ⟹ x i = 2 0 1 6 − k − k
Now, i = 0 ∑ 2 0 1 6 x i 3 = i = 0 ∑ k x i 3 + i = k + 1 ∑ 2 0 1 6 x i 3 = k + ( 2 0 1 6 − k ) x i 3 = k − ( 2 0 1 6 − k ) 2 k 3
Now, using simple calculus maximum value of k = 6 7 2
Using this maximum value of sum is 5 0 4
This is an incomplete solution, because it might be possible to achieve higher value by not setting all positive values as 1.
Log in to reply
if you set values strictly less than 1 then it may be fractions or negative numbers.SO, their cubes will be strictly negative and total positive value will be reduced.
Log in to reply
But if you make some positive value less than 1, then the sum of the positive values becomes smaller. So the negative values also become smaller (in magnitude); this might potentially make the overall sum of cubes to be larger.
The largest fraction that can cancel 1 is 1/2. Two of "-1/2" will cancel one "1" , and in three terms, when cubed, we get 1-2/8=3/4. \therefore the average per 2016 terms =(3/4)/3=1/4.
For 2016 terms, the sum = 2016 * 1/4=504.
I f w e t a k e a f r a c t i o n ( i ) g r e a t e r t h a n o r ( i i ) l e s s t h a n 2 1 , s a y 2 1 ± n 1 = 2 n n ± 2 . I n o r d e r t o c a n c e l " 1 s " , t h i s f r a c t i o n 2 n n ± 2 m u s t h a v e 2 n t e r m s . S o m a x i m u m " 1 s " m u s t b e ( n ± 2 ) t e r m s . S o i n ( 3 n ± 2 ) t e r m s , w e w i l l h a v e ( n ± 2 ) " 1 s " a n d 2 n o f " − 2 n n ± 2 " t e r m s . S u m o f 3 n ± 2 t e r m s = n ± 2 − ( n ± 2 ) = 0 . S u m o f t h e c u b e s f o r 3 n ± 2 t e r m s = n ± 2 − 2 n ∗ ( 2 n n ± 2 ) 3 = ( n ± 2 ) ( 1 − ( 2 n n ± 2 ) 2 = ( n ± 2 ) ( 1 − ( 2 1 ± n 1 ) 2 ) = ( n + 2 ) ( 3 / 4 − n 2 1 ∓ n 1 ) . ∴ a v e r a g e = 3 n ± 2 n ± 2 ∗ ( 3 / 4 − n 2 1 ± n 1 ) < 1 / 4 a v e r a g e f o r 1 / 2 . S a y i f n = ± 4 , a v e r a g e f o r b o t h , ± i s 3 / 1 6 f r o m a b o v e f o r m u l a .
I f w e t a k e a f r a c t i o n ( i ) g r e a t e r t h a n o r ( i i ) l e s s t h a n 2 1 , s a y 2 1 ± n 1 = 2 n n ± 2 . I n o r d e r t o c a n c e l " 1 s " , t h i s f r a c t i o n 2 n n ± 2 m u s t h a v e 2 n t e r m s . S o m a x i m u m " 1 s " m u s t b e ( n ± 2 ) t e r m s . S o i n ( 3 n ± 2 ) t e r m s , w e w i l l h a v e ( n ± 2 ) " 1 s " a n d 2 n o f " − 2 n n ± 2 " t e r m s . S u m o f 3 n ± 2 t e r m s = n ± 2 − ( n ± 2 ) = 0 . S u m o f t h e c u b e s f o r 3 n ± 2 t e r m s = n ± 2 − 2 n ∗ ( 2 n n ± 2 ) 3 = ( n ± 2 ) ( 1 − ( 2 n n ± 2 ) 2 = ( n ± 2 ) ( 1 − ( 2 1 ± n 1 ) 2 ) = ( n + 2 ) ( 3 / 4 − n 2 1 ∓ n 1 ) . ∴ a v e r a g e = 3 n ± 2 n ± 2 ∗ ( 3 / 4 − n 2 1 ± n 1 ) < 1 / 4 a v e r a g e f o r 1 / 2 . S a y i f n = ± 4 , a v e r a g e f o r b o t h , ± i s 3 / 1 6 f r o m a b o v e f o r m u l a .
Problem Loading...
Note Loading...
Set Loading...
Observe that the inequality − 4 x 3 + 3 x + 1 = ( 1 − x ) ( 2 x + 1 ) 2 ≥ 0 holds ∀ x ≤ 1 . Assigning the numbers as x k for some 1 ≤ k ≤ 2 0 1 6 , it is evident that
k = 1 ∑ 2 0 1 6 ( − 4 x k 3 + 3 x k + 1 ) = 2 0 1 6 − 4 k = 1 ∑ 2 0 1 6 x k 3 ≥ 0
⟹ k = 1 ∑ 2 0 1 6 x k 3 ≤ 5 0 4 .
Such a sum is obtained if x k = 1 for 1 ≤ k ≤ 6 7 2 , and x k = 2 − 1 for 6 7 3 ≤ k ≤ 2 0 1 6 .
Hence, the answer is 5 0 4 .