Difference between revisions of "Newton's Inequality"
m (→See Also) |
m (→Proof) |
||
Line 20: | Line 20: | ||
''Proof.'' | ''Proof.'' | ||
− | We consider the [[derivative]] of <math> P(t) = \prod_{i=1}^ | + | We consider the [[derivative]] of <math> P(t) = \prod_{i=1}^m (t+x_i) </math>. The roots of <math>P </math> are <math> -x_1, \ldots, -x_m </math>. Without loss of generality, we assume that the <math>x_i </math> increase as <math>i </math> increases. Now for any <math>i \in [1, m-1] </math>, <math>P'(t) </math> must have a root between <math>x_i </math> and <math>x_{i+1} </math> by [[Rolle's theorem]] if <math>x_i \neq x_{i+1} </math>, and if <math> x_i = x_{i+1} = \cdots = x_{i+k} </math>, then <math>x_{i} </math> is a root of <math>P </math> <math>k+1 </math> times, so it must be a root of <math>P' </math> <math>k </math> times. It follows that <math>P' </math> must have <math>m-1 </math> non-positive, real roots, i.e., for some non-negative reals <math> x'_1, \ldots, x'_{m-1} </math>, |
<center> | <center> | ||
<math> {} | <math> {} |
Revision as of 07:32, 22 January 2018
Contents
Background
For , we define the symmetric sum to be the coefficient of in the polynomial (see Viete's sums). We define the symmetric average to be .
Statement
For non-negative and ,
,
with equality exactly when all the are equal.
Proof
Lemma. For real , there exist real with the same symmetric averages .
Proof. We consider the derivative of . The roots of are . Without loss of generality, we assume that the increase as increases. Now for any , must have a root between and by Rolle's theorem if , and if , then is a root of times, so it must be a root of times. It follows that must have non-positive, real roots, i.e., for some non-negative reals ,
.
It follows that the symmetric sum for is , so the symmetric average .
Thus to prove Newton's theorem, it is sufficient to prove
for any . Since this is a homogenous inequality, we may normalize it so that . The inequality then becomes
.
Expanding the left side, we see that this is
.
But this is clearly equivalent to
,
which holds by the rearrangement inequality.