(1) Prove that on the complex plane, the convex hull of all complex roots of the equation z20+63z+22=0 is larger than π.
(2) Let n be a positive integer, and 1≤k1<k2<⋯<kn be n odd integers. Prove that for any n complex numbers a1,a2,…,an with sum 1 and any complex number w with absolute value at least 1, the equation a1zk1+a2zk2+⋯+anzkn=w has at least one complex root with norm less than or equal to 3n∣w∣.
Solution
(1) We first prove the following.
Lemma: [Gauss-Lucas Theorem] All zeros of the polynomial f′(z) lie in the closure formed by zeros of f(z).
Proof of Lemma: By the fundamental theorem of algebra, we know that f(z) admits a decomposition: f(z)=A(z−z1)α1(z−z2)α2⋯(z−zn)αn. Thus, z1,z2,…,zn are zeros of f′(z) with multiplicities α1−1,α2−1,…,αn−1, respectively. Say Z is any other zero of f′(z), then f(Z)f′(Z)=Z−z1α1+Z−z2α2+⋯+Z−znαn=0. That is, ∣Z−z1∣2α1(Zˉ−zˉ1)+∣Z−z2∣2α2(Zˉ−zˉ2)+⋯+∣Z−zn∣2αn(Zˉ−zˉn)=0. After taking complex conjugations, the equality becomes Z=∣Z−z1∣2α1+∣Z−z2∣2α2+⋯+∣Z−zn∣2αn∣Z−z1∣2α1z1+∣Z−z2∣2α2z2+⋯+∣Z−zn∣2αnzn. Hence Z is a convex linear combination of z1,z2,…,zn. This proves the lemma.
Back to the original problem. Gauss-Lucas Lemma implies that the convex hull of the zeros of the given equation contains the convex hull of the zeros of the following equation 20z19−63=0. This is a regular 19-gon inscribed in a circle of radius (2063)1/19.
The inscribed circle of this regular 19-gon has a radius (2063)1/19cos19π>(2063)1/19(1−21(19π)2)>(2063)1/19(1−21(61)2)=7271(2063)1/19. On the other hand, since (7172)19=(1+711)19<1+7119+18⋅C1927221<3<2063, the radius of the inscribed circle is larger than 1. This proves (1).
(2) We need to prove the following.
(A) If a1,a2,…,an and w satisfy the assumptions, then the minimal norm of the roots of a1zk1+a2zk2+⋯+anzkn=w(∗) is less than or equal to 3mn. This is equivalent to the following.
(B) If a1,a2,…,an and w satisfy the assumptions, then the maximal norm of the roots of wzkn−a1zkn−k1−a2zkn−k2−⋯−an−1zkn−kn−1−an=0,(∗∗) denoted by M0, is greater than or equal to 1/(3mn).
By the lemma in (1), the maximal norm of roots of (∗∗) cannot be less than the maximal norm of roots of wknzkn−1−a1(kn−k1)zkn−k1−1−⋯−an−1(kn−kn−1)zkn−kn−1−1=0.(1) That is, the maximal norm, say M1, of the roots of wknzkn−1−a1(kn−k1)zkn−k1−1−a2(kn−1−k2)zkn−k2−1−⋯−an−1(kn−kn−1)=0.(1′) Beginning with (1′), we repeat the same process for n−s times. It is known that the maximal norm of the roots decreases each time. Therefore, M0≥M1≥⋯≥Mn−s, where Mn−s is the maximal norm of the roots of the equation wknkn−1⋯ks+1zks−a1(kn−k1)(kn−1−k1)⋯(ks+1−k1)zks−k1−⋯−as(kn−ks)⋯(ks+1−ks)=0.((n−s)′)
By Vieta's theorem, we have M0≥M1≥⋯≥Mn−s≥[knkn−1⋯ks+1(kn−ks)⋯(ks+1−ks)⋅was]1/ks. Hence it suffices to show that there exists s∈{1,2,…,n} such that [(kn−ks)(kn−1−ks)⋯(ks+1−ks)knkn−1⋯ks+1]1/ks⋅[asw]1/ks≤3mn. Note that kp−kskp=1+kp−ksks≤(1+kp−ks1)ks. So it remains to show that there exists some s such that [p=s+1∏n(1+kp−ks1)]⋅(∣as∣m)1/ks≤3mn.2 We also make the following observations.
* Since all kj's are odd numbers, kp−ks≥2(p−s). Therefore, 1+kp−ks1≤1+2(p−s)1<p−sp−s+1; * Since ∑j=1naj=1, we must have ∑j=1n∣aj∣≥1. Thus there must be s satisfying ∣as∣≥2s1, or equivalently, ∣as∣1≤2s.
We separate the discussion into the following two cases.
(1) If there is s (allowing s=1), such that ks≥3 and ∣as∣≥2s1, then (note that s≤2s−1≤ks) [p=s+1∏n(1+kp−ks1)]⋅(∣as∣m)1/ks<[p=s+1∏np−sp−s+1]⋅m1/ks⋅2s/ks<n⋅3m⋅2<3mn.
(2) Otherwise, we must have k1=1 and ∣a1∣≥21. Hence [p=2∏n(1+kp−k11)]⋅∣a1∣m≤[p=2∏n2(p−1)2(p−1)+1]⋅m⋅2<k=1∏2n−2kk+1⋅63⋅2=2n−1⋅m⋅2<3mn.
This completes the proof.
Want a route through all this instead of an archive? The track
puts 2,000 problems in a working order, from AMC 10 level to the IMO shortlist.
Source: MathNet,
licensed CC-BY-4.0.
Statement and solution reproduced as published; topic and difficulty added by this site.