Find the largest possible product n ⋅ m of two distinct positive integers n and m , such that the equations ( x + 1 ) n = x n + 1 and ( x + 1 ) m = x m + 1 have the same set of complex roots.
Details and assumptions
The multiplicities of the roots can, of course, be different, but each root of one equation is a root of the other.
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.
Holy god... for some reason I interpreted the question as "find the two largest m , n such that the equation ( x + 1 ) m has the same roots as x m + 1 . Needless to say, I was very confused...
The algebra at the end is even simpler if we check the first three coefficients in the identity ( x − 1 ) f n ( x ) = n − 2 n ( x 3 − 1 ) f n − 2 ( x )
Very nice.
I did exactly the same as you up until I got to m = n − 2 = 6 k − 1 , and noting that k = 1 works, but as so often happens with these problems, it was far easier to guess that k = 1 was the solution than to do the work of proving that this was the case.
Very nice!
Yet another way to finish the problem after the identity f n ( x ) = n − 2 n ( x 2 + x + 1 ) f n − 2 ( x ) is established, is to plug in x = 1 and use some inequalities.
Log in to reply
I guess we don't need inequalities. Putting x = 1 yields ( n − 2 ) ( 2 n − 2 ) = 3 n ( 2 n − 2 − 2 ) , so that ( n − 8 ) 2 n − 2 + 4 n + 4 = 0 where n ≥ 4 . Thus n ≤ 7 ; since n ≡ 1 modulo 6 we are done.
Is (5,7) the only m and n that work then?
By substituting integers into n and m, we'll be able to get n=5 and m=7. Therefore the product of n and m is 35.
Care to explain a bit more?
It's really not hard to do! You could just pull up Wolfram|Alpha, but it's not too hard to hack out. When you do so, you see that m = 7 and m = 5 have the same roots. I didn't check for any others, but it turns out that 3 5 is the correct answer.
Problem Loading...
Note Loading...
Set Loading...
Let S n be the set of roots of f n ( x ) = ( 1 + x ) n − x n − 1 = 0 . If m < n and S m = S n , then S n must contain at least one repeated root. Suppose that u ∈ C is a repeated root in S n . Then f n ( u ) = ( 1 + u ) n − u n − 1 = 0 and f n ′ ( u ) = n ( 1 + u ) n − 1 − n u n − 1 = 0 , so that 1 + u n = ( 1 + u ) n = ( 1 + u ) u n − 1 = u n − 1 + u n and hence ( 1 + u ) n − 1 = u n − 1 = 1 . Since u and 1 + u are ( n − 1 ) st roots of unity (and hence complex numbers of modulus 1 ) which differ by 1 , we deduce that u = − 2 1 ± 2 1 i 3 = ω , ω 2 1 + u = 2 1 ± 2 1 i 3 = − ω 2 , − ω where ω is the primitive third root of unity. Since both u and 1 + u are ( n − 1 ) st roots of unity, we deduce that n − 1 is a multiple of 6 .
Thus ω and ω 2 are the only possible roots of S n that can be repeated. Since f n ′ ′ ( x ) = n ( n − 1 ) [ ( 1 + x ) n − 2 − x n − 2 ] it is clear that f n ′ ′ ( ω ) = 0 , and so ω cannot be more than a double root in S n (and similarly for ω 2 ). Since f n ( x ) has real coefficients, we see that both ω and ω 2 occur as double roots of S n . Thus we deduce that S n contains n − 2 elements.
Hence we deduce that S m must have n − 2 roots, including ω and ω 2 . If these two were double roots, then m − 1 would be a multiple of 6 , as for n , and S m would contain m − 2 roots. The only way this would be possible would be if m = n , which is not the case. This we deduce that S m contains no double roots, and hence that m = n − 2 . But this would imply (since f n ( x ) has leading term n x n − 1 ) that f n ( x ) = n − 2 n ( x 2 + x + 1 ) f n − 2 ( x ) Using the Binomial Theorem and multiplying out the first few terms, f n ( x ) f n − 2 ( x ) ( x 2 + x + 1 ) f n − 2 ( x ) = = = 1 n x n − 1 + 2 n ( n − 1 ) x n − 2 + 6 n ( n − 1 ) ( n − 2 ) x n − 3 + ⋯ 1 n − 2 x n − 3 + 2 ( n − 2 ) ( n − 3 ) x n − 4 + 6 ( n − 2 ) ( n − 3 ) ( n − 4 ) x n − 5 + ⋯ 1 n − 2 x n − 1 + 2 ( n − 1 ) ( n − 2 ) x n − 2 + 6 ( n − 2 ) ( n 2 − 4 n + 9 ) x n − 3 + ⋯ and so, matching the coefficients of x n − 3 , 6 1 n ( n − 1 ) ( n − 2 ) n 2 − 3 n + 2 = ( n − 1 ) ( n − 2 ) = = n − 2 n × 6 1 ( n − 2 ) ( n 2 − 4 n + 9 ) n 2 − 4 n + 9 and hence n = 7 . We can calculate that ( 1 + x ) 5 − x 5 − 1 = 5 x ( x + 1 ) ( x 2 + x + 1 ) , while ( 1 + x ) 7 − x 7 − 1 = 7 x ( x + 1 ) ( x 2 + x + 1 ) 2 , and hence the case m = 5 , n = 7 is possible. Thus the largest, indeed the only, value of m × n where m and n are distinct and S m = S n is 5 × 7 = 3 5 .
Once we have limited the possible values of double/repeated roots of these polynomials, the rest is relatively straightforward. Note that the coefficient of x n − 3 is the first one which is not automatically the same on both sides of the identity f n ( x ) = n − 2 n ( x 2 + x + 1 ) f n − 2 ( x ) .