Operators preserving real zeros (Laguerre-Pólya theory)

[49/84%]
Show
LaTeX

Let pp be a nonconstant polynomial with real coefficients and only real roots. Prove that for each real number rr, the polynomial p−rp′p - rp' has only real roots.

?

All the roots of the polynomial a0xn+a1xn−1+…+ana_{0}x^{n}+a_{1}x^{n-1}+\ldots +a_{n} are real. Prove that all the roots of the polynomial

(n+1)sa0xn+nsa1xn−1+…+an,s an integer (n+1)^{s}a_{0}x^{n}+n^{s}a_{1}x^{n-1}+\ldots +a_{n}, \quad s \text{ an integer}

are real as well.

?

All the roots of the polynomial a0xn+a1xn−1+…+ana_{0}x^{n}+a_{1}x^{n-1}+\ldots +a_{n} are real. Prove that the roots of the equation

asxk+Ck1as+1xk−1+Ck2as+2xk−2+…+as+k=0,s+k⩽n a_{s}x^{k}+C_{k}^{1}a_{s+1}x^{k-1}+C_{k}^{2}a_{s+2}x^{k-2}+\ldots +a_{s+k}=0, \qquad s+k \leqslant n

are real as well.

?

All the roots of the polynomial f(x)f(x) are real. Prove that the roots of the polynomial f(x)+λf′(x)f(x)+\lambda f'(x) are also real, if λ\lambda is real.

?

Under the same condition (see problem 2539), suppose the polynomial f(x)+mf(x)+m has kk imaginary1 roots. Prove that the equation f′(x)2−f(x)f′′(x)−mf′′′(x)=0f'(x)^{2}-f(x)f''(x)-mf'''(x)=0 has at least kk real roots.

Footnotes

  1. Imaginary numbers are complex numbers with nonzero imaginary part (i.e., not real; not to be confused with purely imaginary numbers). ↩

?

Prove the theorem: if all the roots of the polynomial f(x)f(x) are real, then for λ>0\lambda > 0 all the roots of the polynomial xf′(x)+λf(x)xf'(x)+\lambda f(x) are real as well. (Laguerre.)

?

Prove the theorem: if all the roots of the polynomial f(x)=a+a1x+a2x2+…+anxnf(x)=a+a_{1}x+a_{2}x^{2}+\ldots +a_{n}x^{n} are real, and all the roots of the polynomial φ(ν)\varphi (\nu ) are real and nonpositive, then all the roots of the polynomial F(x)=∑aνφ(ν)xνF(x)=\sum a_{\nu }\varphi (\nu )x^{\nu } are real as well. (Laguerre.)

?

Prove the theorem: if all the roots of the polynomial f(x)=a+a1x+a2x2+…+anxnf(x)=a+a_{1}x+a_{2}x^{2}+\ldots +a_{n}x^{n} are real, then all the roots of the polynomial F(x)=∑aνψ(ν)xνF(x)=\sum a_{\nu }\psi (\nu )x^{\nu } are real as well, where ψ(ν)=eAν∏(1+νaν)e−νaν\psi (\nu )=e^{A\nu }\prod \left(1+\dfrac {\nu }{a_{\nu }}\right)e^{-\frac{\nu }{a_{\nu }}}, the numbers aνa_{\nu } are positive, and AA is real. (Laguerre.)

?

If the function Φ(x)=f(x)+af′(x)+a2f′′(x)+…+anf(n)(x)\Phi (x)=f(x)+af'(x)+a^{2}f''(x)+\ldots +a^{n}f^{(n)}(x), where f(x)f(x) is a polynomial of degree nn, has real roots, then so does the polynomial f(x)f(x).

?

We assume that all the zeros of the polynomial

P(z)=a0zn+a1zn−1+⋯+an−1z+an P(z) = a_{0}z^{n} + a_{1}z^{n-1} + \cdots + a_{n-1}z + a_{n}

are in the upper half-plane Im(z)>0\mathrm{Im}\left(z\right) > 0. Let aν=αν+iβνa_{\nu } = \alpha_{\nu } + i\beta_{\nu }, αν,βν\alpha_{\nu }, \beta_{\nu } real, ν=0,1,2,…,n\nu = 0, 1, 2, \ldots , n, and

U(z)=α0zn+α1zn−1+⋯+αn−1z+αn, U(z) = \alpha _{0}z^{n} + \alpha _{1}z^{n-1} + \cdots + \alpha _{n-1}z + \alpha _{n}, V(z)=β0zn+β1zn−1+⋯+βn−1z+βn. V(z) = \beta _{0}z^{n} + \beta _{1}z^{n-1} + \cdots + \beta _{n-1}z + \beta _{n}.

Then the polynomials U(z)U(z) and V(z)V(z) have only real zeros.

?

If a polynomial f(z)f(z) whose coefficients are all real has only real zeros then this is true also for its derivative f′(z)f'(z). If f(z)f(z) has complex zeros then they appear in pairs, the two zeros forming a pair are mirror images to each other with respect to the real axis; they are complex conjugates. We draw all those disks the ``vertical'' diameters of which are the line segments connecting the conjugate zeros of such pairs. If f′(z)f'(z) has any complex zeros they lie in these disks.

?
Remark.
?

Examine the imaginary part of f′(z)f(z)\frac{f'(z)}{f(z)}.

The derivative of a polynomial has not more non-real zeros than the polynomial itself.

?

Let μ,ν\mu , \nu be integers satisfying 0≤μ<μ+ν≤n0 \leq \mu < \mu +\nu \leq n. None of the polynomials

aμ+(ν1)aμ+1x+(ν2)aμ+2x2+⋯+(νν−1)aμ+ν−1xν−1+aμ+νxν a_{\mu } + \binom {\nu }{1} a_{\mu +1} x + \binom {\nu }{2} a_{\mu +2} x^2 + \cdots + \binom {\nu }{\nu -1} a_{\mu +\nu -1} x^{\nu -1} + a_{\mu +\nu } x^{\nu }

has more nonreal zeros than the polynomial

a0+(n1)a1x+(n2)a2x2+⋯+(nn−1)an−1xn−1+anxn. a_0 + \binom {n}{1} a_1 x + \binom {n}{2} a_2 x^2 + \cdots + \binom {n}{n-1} a_{n-1} x^{n-1} + a_n x^n.

(Generalization of 53.)

?

Let α\alpha be real and P(x)P(x) a polynomial. The polynomial αP(x)+P′(x)\alpha P(x)+P'(x) has not more non-real zeros than the polynomial P(x)P(x) itself. (Generalization of 53.)

?

Assume that the equation

a0+a1x+a2x2+⋯+anxn=0 a_0+a_1x+a_2x^2+\cdots +a_nx^n=0

has only real roots. Then the polynomial

a0P(x)+a1P′(x)+⋯+anP(n)(x) a_0 P(x) + a_1 P'(x) + \cdots + a_n P^{(n)}(x)

has not more non-real zeros than the polynomial P(x)P(x) itself.

?
P(x)−P′′(x)1!+P(IV)(x)2!−P(VI)(x)3!+⋯ P(x) - \frac{P''(x)}{1!} + \frac{P^{(\mathrm{IV})}(x)}{2!} - \frac{P^{(\mathrm{VI})}(x)}{3!} + \cdots

has not more non-real zeros than the polynomial P(x)P(x) itself.

?

If the equation

a0+a1x+a2x2+⋯+anxn=0 a_0+a_1x+a_2x^2+\cdots +a_nx^n=0

has only real roots, then the following equation

a0+a11!x+a22!x2+⋯+ann!xn=0 a_0+\frac{a_1}{1!}x+\frac{a_2}{2!}x^2+\cdots +\frac{a_n}{n!}x^n=0

has also only real roots.

?

Let P(x)P(x) be a polynomial of nnth degree, and let the real number α\alpha lie outside the interval [−n,0][-n,0]. Then αP(x)+xP′(x)\alpha P(x)+xP'(x) has not more non-real zeros than P(x)P(x). (A generalization of 53 different from 60 and 62.)

?

Let Q(x)Q(x) be a polynomial whose zeros are all real and lie outside the interval [0,n][0,n]. Then the equation

a0Q(0)+a1Q(1)x+a2Q(2)x2+⋯+anQ(n)xn=0 a_0Q(0)+a_1Q(1)x+a_2Q(2)x^2+\cdots +a_nQ(n)x^n=0

has not more non-real roots than the equation

a0+a1x+a2x2+⋯+anxn=0. a_0+a_1x+a_2x^2+\cdots +a_nx^n=0.
?

Let 0<q<10<q<1. The equation

a0+a1qx+a2q4x2+⋯+anqn2xn=0 a_0+a_1qx+a_2q^4x^2+\cdots +a_nq^{n^2}x^n=0

has not more non-real roots than the equation

a0+a1x+a2x2+⋯+anxn=0. a_0+a_1x+a_2x^2+\cdots +a_nx^n=0.
?

Let q>0q > 0. The equation

2a0+(q+q−1)a1x+(q2+q−2)a2x2+⋯+(qn+q−n)anxn=0 2a_0 + (q+q^{-1}) a_1 x + \left(q^{\sqrt{2}}+q^{-\sqrt{2}}\right) a_2 x^2 + \cdots + \left(q^{\sqrt{n}}+q^{-\sqrt{n}}\right) a_n x^n = 0

has not more non-real roots than the equation

a0+a1x+a2x2+⋯+anxn=0. a_0+a_1x+a_2x^2+\cdots +a_nx^n=0.
?

Let 0<a<b0<a<b. If f(x)f(x) vanishes at n+1n+1 points of the interval [a,b][a,b] and if all the zeros of the polynomial with real coefficients a0+a1x+a2x2+⋯+anxna_0+a_1x+a_2x^2+\cdots +a_nx^n are real, then at an interior point ξ\xi of [a,b][a,b]

a0f(ξ)+a1f′(ξ)+a2f′′(ξ)+⋯+anf(n)(ξ)=0.[63.] a_0f(\xi )+a_1f'(\xi )+a_2f''(\xi )+\cdots +a_nf^{(n)}(\xi )=0. \qquad [63.]
?

Deduce from

lim⁡n→∞(1+izn)n=cos⁡z+isin⁡z, \lim _{n \to \infty } \left(1+\frac{iz}{n}\right)^n = \cos z+i\sin z,

that cos⁡z\cos z and sin⁡z\sin z have no non-real zeros.

?

Deduce from

lim⁡n→∞(1−z2n)n=e−z2, \lim _{n \to \infty } \left(1-\frac{z^2}{n}\right)^n = e^{-z^2},

that no derivative of e−z2e^{-z^2} has non-real zeros.

?

The Bessel function of order 00

J0(z)=1−11!1!(z2)2+12!2!(z2)4−13!3!(z2)6+⋯ J_0(z) = 1-\frac{1}{1!1!}\left(\frac{z}{2}\right)^2+\frac{1}{2!2!}\left(\frac{z}{2}\right)^4-\frac{1}{3!3!}\left(\frac{z}{2}\right)^6+\cdots

has no non-real zeros. One can obtain from the four different representations

  1. J0(z)=lim⁡n→∞(−1)n(1+z24n2)nPn(z2−4n2z2+4n2)J_0(z) = \lim \limits_{n \to \infty } (-1)^n\left(1+\dfrac {z^2}{4n^2}\right)^n P_n\left(\dfrac {z^2-4n^2}{z^2+4n^2}\right) (problem 85 of Part Six),

  2. J0(z)=lim⁡n→∞Ln(z24n)J_0(z) = \lim \limits_{n \to \infty } L_n\left(\dfrac {z^2}{4n}\right) (problem 99(b) of Part Six’s solution),

  3. J0(z)=12π∫−ππeizsin⁡ϑ dϑJ_0(z) = \dfrac {1}{2\pi }\displaystyle \int_{-\pi }^{\pi } e^{iz\sin \vartheta }\, d\vartheta (problems 148, 56 of Part Three),

  4. J0(z)=2π∫01cos⁡zt1−t2 dtJ_0(z) = \dfrac {2}{\pi }\displaystyle \int_0^1 \dfrac {\cos zt}{\sqrt{1-t^2}}\, dt (problem 205 of Part Three) four different proofs of this result.

?

Let qq be an integer, q≥2q \geq 2. The entire function

F(z)=1+zq!+z2(2q)!+z3(3q)!+⋯ F(z) = 1+\frac{z}{q!}+\frac{z^2}{(2q)!}+\frac{z^3}{(3q)!}+\cdots

has no non-real zeros. One can obtain from 59 two different proofs of this result.

?

Let α≥0\alpha \geq 0, 0<α1≤α2≤α3≤⋯0<\alpha_1 \leq \alpha_2 \leq \alpha_3 \leq \cdots, and 1/α1+1/α2+1/α3+⋯1/\alpha_1+1/\alpha_2+1/\alpha_3+\cdots be convergent. The transcendental entire function

g(z)=e−αz(1−zα1)(1−zα2)(1−zα3)⋯ g(z) = e^{-\alpha z}\left(1-\frac{z}{\alpha _1}\right)\left(1-\frac{z}{\alpha _2}\right)\left(1-\frac{z}{\alpha _3}\right)\cdots

may be represented as the limit of a sequence of polynomials with only real positive zeros.

?

If

g(z)=1+a11!z+a22!z2+a33!z3+⋯ , g(z) = 1+\frac{a_1}{1!}z+\frac{a_2}{2!}z^2+\frac{a_3}{3!}z^3+\cdots ,

then the polynomials

1+(n1)a1z+(n2)a2z2+⋯+(nn−1)an−1zn−1+anzn,n=1,2,3,… 1+\binom {n}{1}a_1z+\binom {n}{2}a_2z^2+\cdots +\binom {n}{n-1}a_{n-1}z^{n-1}+a_nz^n, \qquad n=1,2,3,\ldots

have only real, positive zeros [63].

?

In the interval 0<x≤α10<x \leq \alpha_1

g(x)<1 g(x)<1

and furthermore we have there more generally

1+a11!x+a22!x2+⋯+a2m−1(2m−1)!x2m−1<g(x) 1+\frac{a_1}{1!}x+\frac{a_2}{2!}x^2+\cdots +\frac{a_{2m-1}}{(2m-1)!}x^{2m-1} < g(x) <1+a11!x+a22!x2+⋯+a2m(2m)!x2m,m=1,2,3,…. < 1+\frac{a_1}{1!}x+\frac{a_2}{2!}x^2+\cdots +\frac{a_{2m}}{(2m)!}x^{2m}, \qquad m=1,2,3,\ldots .

[Thus g(x)g(x) is enveloped by its Maclaurin series for 0<x≤α10<x \leq \alpha_1; 55, 72.]

?

Assume α<1\alpha <1. The entire function of zz defined by the integral

∫0∞e−t2g(−t2)cos⁡zt dt \int _0^{\infty } e^{-t^2}g(-t^2)\cos zt\, dt

has only real zeros [63].

?

Let α,β,β1,β2,β3,…\alpha , \beta , \beta_1, \beta_2, \beta_3, \ldots be real, α≥0\alpha \geq 0, βν≠0\beta_{\nu } \neq 0, and the series 1/β12+1/β22+1/β32+⋯1/\beta_1^2+1/\beta_2^2+1/\beta_3^2+\cdots convergent. The transcendental entire function

G(z)=e−αz2+βz(1−zβ1)ezβ1(1−zβ2)ezβ2⋯ G(z) = e^{-\alpha z^2+\beta z}\left(1-\frac{z}{\beta _1}\right)e^{\frac{z}{\beta _1}}\left(1-\frac{z}{\beta _2}\right)e^{\frac{z}{\beta _2}}\cdots

may be represented as the limit of a sequence of polynomials with only real zeros.

?

If

G(z)=1+b11!z+b22!z2+b33!z3+⋯ , G(z) = 1+\frac{b_1}{1!}z+\frac{b_2}{2!}z^2+\frac{b_3}{3!}z^3+\cdots ,

then

bm2+bm+12>0,m=1,2,3,….[49.] b_m^2+b_{m+1}^2>0, \qquad m=1,2,3,\ldots . \qquad [49.]
?

If G(z)G(z) has no positive zeros and a0,a1,a2,…,ana_0, a_1, a_2, \ldots , a_n are real numbers, then the polynomial

a0G(0)+a1G(1)z+a2G(2)z2+⋯+anG(n)zn a_0G(0)+a_1G(1)z+a_2G(2)z^2+\cdots +a_nG(n)z^n

can have not more non-real zeros than the polynomial

a0+a1z+a2z2+⋯+anzn. a_0+a_1z+a_2z^2+\cdots +a_nz^n.

[68, 69 are special cases.]

?

Prove from 167 that the Bessel function J0(z)J_0(z) has no non-real zeros (problem 31 of Part Two).

?

Prove from 167 that the entire functions considered in 160 have no non-real zeros.

?

Let α\alpha be an even integer, α≥2\alpha \geq 2. Then the integral

∫0∞e−tαcos⁡zt dt=Fα(z) \int _0^{\infty } e^{-t^{\alpha }}\cos zt\, dt = F_{\alpha }(z)

represents an entire function that has no non-real zeros. [167.]

?

Let a≥2a \geq 2. The entire function

F(z)=1+za+z2a4+z3a9+⋯+znan2+⋯ F(z) = 1+\frac{z}{a}+\frac{z^2}{a^4}+\frac{z^3}{a^9}+\cdots +\frac{z^n}{a^{n^2}}+\cdots

as well as all the partial sums of its power series, has only real negative simple zeros. (Problem 200 of Part Three.)

?

Let the degree of the polynomial H(x)H(x) be ≥3\geq 3. If

F(x)=eH(x), F(x)=e^{H(x)},

then at least one of the two functions dF/dxdF/dx and d2F/dx2d^2F/dx^2 has not only real zeros.

?

Let the polynomial f(x)f(x) be such that, if P(x)P(x) is an arbitrary polynomial, P(x)f(x)+P′(x)P(x)f(x)+P'(x) has not more non-real zeros than P(x)P(x).

For this it is necessary and sufficient, that f(x)=a−bxf(x)=a-bx, where aa is arbitrary, b≥0b \geq 0.

?

Let α1,α2,…,αs\alpha_1, \alpha_2, \ldots , \alpha_s denote the zeros of G(z)G(z) in [0,m][0,m],

0≤α1<α2<⋯<αs≤m, 0 \leq \alpha _1<\alpha _2<\cdots <\alpha _s \leq m,

and set

G(z)=(z−α1)(z−α2)…(z−αs)G∗(z). G(z) = (z-\alpha _1)(z-\alpha _2)\ldots (z-\alpha _s)G^*(z).

Then argue as follows:

(1)

a0G∗(0)+a1G∗(1)z+⋯+anG∗(n)zn a_0G^*(0)+a_1G^*(1)z+\cdots +a_nG^*(n)z^n

has precisely mm changes of sign (by 47 and the definition of G∗(z)G^*(z)) and precisely mm positive zeros (by a slight modification of 167---make use of 67 more completely).

(2)

a0G∗(0)(0−α1)+a1G∗(1)(1−α1)z+⋯+amG∗(m)(m−α1)zm a_0G^*(0)(0-\alpha _1)+a_1G^*(1)(1-\alpha _1)z+\cdots +a_mG^*(m)(m-\alpha _1)z^m

has precisely m−1m-1 changes of sign (attention to the various possible cases) and so by 36 at most m−1m-1 positive zeros. Yet it has (by a slight modification of solution 66) at least m−1m-1, and so finally exactly m−1m-1, positive zeros;

(3) Introduce successively α2,α3,…,αs\alpha_2, \alpha_3, \ldots , \alpha_s and repeat appropriately the arguments of (2) (attention at each step to the changes wrought by the preceding steps) to arrive at the desired proposition.

?

We make the following assumptions: The entire function

g(z)=a0+a11!z+a22!z2+a33!z3+⋯ g(z) = a_0+\frac{a_1}{1!}z+\frac{a_2}{2!}z^2+\frac{a_3}{3!}z^3+\cdots

is a limit of polynomials with only positive zeros, but not itself a polynomial [161].

The entire function G(z)G(z) is either defined as in 165 or it is equal to the function defined in 165 multiplied by zz; it has precisely ss non-negative zeros which are simple and the distance between any two of them is not less than 11. (We obtain the present G(z)G(z) from the G(z)G(z) of 171.1 by making mm arbitrarily large.)

Then the entire function

a0G(0)+a1G(1)1!z+a2G(2)2!z2+⋯ a_0G(0)+\frac{a_1G(1)}{1!}z+\frac{a_2G(2)}{2!}z^2+\cdots

has precisely ss non-positive zeros.

?

If ν≥0\nu \geq 0 the entire function

zν2J−ν(2z1/2)=∑n=0∞(−z)nn!1Γ(n+1−ν) z^{\frac{\nu }{2}}J_{-\nu }(2z^{1/2}) = \sum _{n=0}^{\infty } \frac{(-z)^n}{n!}\frac{1}{\Gamma (n+1-\nu )}

has precisely [ν][\nu ] non-positive zeros. (This contains the result of 159.)

?

Let k,l,mk, l, m and qq be non-negative integers,

0≤k<l,k+ql≤m<k+(q+1)l. 0 \leq k<l, \qquad k+ql \leq m<k+(q+1)l.

If all the zeros of the polynomial

b0+b1z+b2z2+⋯+bmzm b_0+b_1z+b_2z^2+\cdots +b_mz^m

are negative, then the zeros of

bk+bk+lz+bk+2lz2+⋯+bk+qlzq b_k+b_{k+l}z+b_{k+2l}z^2+\cdots +b_{k+ql}z^q

are also all negative. (This implies, for n→∞n \to \infty, the result of 160.)

?

If all the zeros of the polynomial

b0+b11!z+b22!z2+⋯+bmm!zm b_0+\frac{b_1}{1!}z+\frac{b_2}{2!}z^2+\cdots +\frac{b_m}{m!}z^m

are negative, then the zeros of

bk+bk+l1!z+bk+2l2!z2+⋯+bk+qlq!zq b_k+\frac{b_{k+l}}{1!}z+\frac{b_{k+2l}}{2!}z^2+\cdots +\frac{b_{k+ql}}{q!}z^q

are also all negative; k,l,mk, l, m and qq are subject to the same conditions as in 171.4.

?

Consider the polynomials P(z),Q(z),R(z)P(z), Q(z), R(z) and

S(z)=a0zn+a1zn−1+⋯+aνzn−ν+⋯+an. S(z) = a_0z^n+a_1z^{n-1}+\cdots +a_{\nu }z^{n-\nu }+\cdots +a_n.

Assume that

P(z) has real coefficients, P(z) \text{ has real coefficients}, Q(z) has only negative zeros, Q(z) \text{ has only negative zeros}, R(z) has only real zeros, R(z) \text{ has only real zeros}, S(z) has only zeros whose absolute value equals 1. S(z) \text{ has only zeros whose absolute value equals } 1.

Then

a0P(z+ni)+⋯+aνP(z+(n−2ν)i)+⋯+anP(z−ni) a_0P(z+ni)+\cdots +a_{\nu }P(z+(n-2\nu )i)+\cdots +a_nP(z-ni)

has no more non-real zeros than P(z)P(z),

a0Q(niz)+⋯+aνQ((n−2ν)iz)+⋯+anQ(−niz) a_0Q(niz)+\cdots +a_{\nu }Q((n-2\nu )iz)+\cdots +a_nQ(-niz)

has only real zeros, and

a0R(ni)zn+⋯+aνR((n−2ν)i)zn−ν+⋯+anR(−ni) a_0R(ni)z^n+\cdots +a_{\nu }R((n-2\nu )i)z^{n-\nu }+\cdots +a_nR(-ni)

has only zeros whose absolute value equals 11.

?

Assume further that a0≠0a_0 \neq 0 and consider the power series

1a0+a1x+a2x2+⋯+anxn=b0+b1x+b2x2+⋯ . \frac{1}{a_0+a_1x+a_2x^2+\cdots +a_nx^n} = b_0+b_1x+b_2x^2+\cdots .

Then the polynomial

b0P(x)+b1P′(x)+b2P′′(x)+⋯+bkP(k)(x)+⋯ b_0P(x)+b_1P'(x)+b_2P''(x)+\cdots +b_kP^{(k)}(x)+\cdots

has not less non-real zeros than the polynomial P(x)P(x) itself.

?

Let G(z)G(z) be an entire function that takes real values for real zz, with G(0)=1G(0) = 1, and

−G′(z)G(z)=s1+s2z+s3z2+⋯+smzm−1+⋯ . -\frac{G'(z)}{G(z)} = s_{1} + s_{2} z + s_{3} z^{2} + \cdots + s_{m} z^{m-1} + \cdots .

If G(z)G(z) has infinitely many zeros, all real, and is of the form given in Problem 165 of Part V, then, for n=1,2,3,…n = 1, 2, 3, \ldots,

∣s2s3s4…sn+1s3s4s5…sn+2s4s5s6…sn+3\multicolumn5c\dotfillsn+1sn+2sn+3…s2n∣>0. \begin{vmatrix} s_{2} & s_{3} & s_{4} & \ldots & s_{n+1} \\ s_{3} & s_{4} & s_{5} & \ldots & s_{n+2} \\ s_{4} & s_{5} & s_{6} & \ldots & s_{n+3} \\ \multicolumn {5}{c}{\dotfill } \\ s_{n+1} & s_{n+2} & s_{n+3} & \ldots & s_{2n} \end{vmatrix} > 0.

(If G(z)G(z) of the form in Problem 165 of Part V has only a finite number of zeros, then only a finite number of these recurrent determinants can be ≠0\neq 0 [Problem 27].)

?

Let G(z)G(z) be an entire function of the form given in Problem 165 of Part V, but not of the form eβze^{\beta z}, and

1G(z)=c0+c11!z+c22!z2+c33!z3+⋯ . \frac{1}{G(z)} = c_{0} + \frac{c_{1}}{1!} z + \frac{c_{2}}{2!} z^{2} + \frac{c_{3}}{3!} z^{3} + \cdots .

Then, for n=0,1,2,…n = 0, 1, 2, \ldots,

∣c0c1c2…cnc1c2c3…cn+1c2c3c4…cn+2\multicolumn5c\dotfillcncn+1cn+2…c2n∣>0. \begin{vmatrix} c_{0} & c_{1} & c_{2} & \ldots & c_{n} \\ c_{1} & c_{2} & c_{3} & \ldots & c_{n+1} \\ c_{2} & c_{3} & c_{4} & \ldots & c_{n+2} \\ \multicolumn {5}{c}{\dotfill } \\ c_{n} & c_{n+1} & c_{n+2} & \ldots & c_{2n} \end{vmatrix} > 0.
?

Let aa be an arbitrary real number. Prove that if the polynomial P(z)=zn+a1zn−1+…+anP(z) = z^{n}+a_{1} z^{n-1}+\ldots +a_{n} has nn real roots, then the polynomial Q(z)=P(z+ia)+P(z−ia)Q(z) = P(z+i a)+P(z-i a) also has nn real roots.

?

References

[deSi04] de Souza P.N., Silva J. Berkeley Problems in Mathematics. 3rd ed., Springer, 2004.
[ГюКу03] Гюнтер Н.М., Кузьмин Р.О. Сборник задач по высшей математике. 13-е изд., Издательство «Лань», 2003.
[PóSz98] Pólya G., Szegő G. Problems and Theorems in Analysis. Springer, 1998.
[ШПК15] Шабунин М.И., Половинкин Е.С., Карлов М.И. Сборник задач по теории функций комплексного переменного. 4-е изд., БИНОМ. Лаборатория знаний, 2015.