The function satisfies the condition for all . Find the sum of all possible values of .
Let this sum be . Submit your answer as the sum of digits of .
Assumption : is the set of all positive integers.
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.
I claim that all of 1 , 2 , 3 , … , 2 0 0 8 can be the value of f ( 2 0 0 7 ) , and no others. This means the sum is 1 + 2 + 3 + … + 2 0 0 8 = 2 0 1 7 0 3 6 and thus the sum of its digits is 2 + 0 + 1 + 7 + 0 + 3 + 6 = 1 9 .
First, the proof that all of them work. The following f works, for k = 0 , 1 , 2 , … , 2 0 0 7 :
f ( n ) = k ⌊ 2 0 0 7 n ⌋ + 1
If k < 2 0 0 7 , then we can verify that f ( n ) ≤ n for all n . Also, we have the easy property ⌊ x + y ⌋ ≥ ⌊ x ⌋ + ⌊ y ⌋ for all reals x , y . Thus,
f ( m + n ) = k ⌊ 2 0 0 7 m + n ⌋ + 1 ≥ k ( ⌊ 2 0 0 7 m ⌋ + ⌊ 2 0 0 7 n ⌋ ) + 1 ≥ k ( ⌊ 2 0 0 7 m ⌋ + ⌊ 2 0 0 7 f ( n ) ⌋ ) + 1 = k ⌊ 2 0 0 7 m ⌋ + 1 + k ⌊ 2 0 0 7 f ( n ) ⌋ + 1 − 1 = f ( m ) + f ( f ( n ) ) − 1
For k = 2 0 0 7 , we don't have the property that f ( n ) ≤ n for all n . We do still retain the property that ⌊ 2 0 0 7 n ⌋ = ⌊ 2 0 0 7 f ( n ) ⌋ , though:
⌊ 2 0 0 7 f ( n ) ⌋ = ⌊ 2 0 0 7 2 0 0 7 ⌊ n / 2 0 0 7 ⌋ + 1 ⌋ = ⌊ ⌊ 2 0 0 7 n ⌋ + 2 0 0 7 1 ⌋ = ⌊ 2 0 0 7 n ⌋ + ⌊ 2 0 0 7 1 ⌋ = ⌊ 2 0 0 7 n ⌋
Thus we can approach exactly as the above.
Now that we have proven those f work, we simply plug in n = 2 0 0 7 to get f ( 2 0 0 7 ) = k + 1 . Since k ranges from 0 to 2007, f ( 2 0 0 7 ) ranges from 1 to 2008.
To prove that no other value of f ( 2 0 0 7 ) is possible, it suffices to prove that f ( n ) ≤ n + 1 for all n , so f ( 2 0 0 7 ) ≤ 2 0 0 8 .
First, note that f ( f ( n ) ) ≥ 1 (because f has codomain the naturals), so f ( m + n ) ≥ f ( m ) + f ( f ( n ) ) − 1 ≥ f ( m ) + 1 − 1 = f ( m ) . In other words, f is non-decreasing.
Now, suppose there exists some a such that f ( a ) ≥ a + 2 . We claim that f ( k a ) ≥ k ( a + 1 ) + 1 for all positive integer k by induction on k . The base case k = 1 is just the assumption, and for the inductive step,
f ( ( k + 1 ) a ) = f ( k a + a ) ≥ f ( k a ) + f ( f ( a ) ) − 1 ≥ k ( a + 1 ) + 1 + f ( a + 2 ) − 1 ≥ k ( a + 1 ) + ( a + 2 ) = ( k + 1 ) ( a + 1 ) + 1
In particular, for k = a , we have f ( a 2 ) ≥ a 2 + a + 1 . Thus,
f ( a + a 2 ) ≥ f ( a ) + f ( f ( a 2 ) ) − 1 ≥ ( a + 2 ) + f ( a 2 + a + 1 ) − 1 ≥ ( a + 1 ) + f ( a 2 + a )
In other words, a ≤ − 1 , contradiction. Thus there can be no such a , proving that f ( n ) ≤ n + 1 for all n . This completes the proof.