olymmath

SCHWEITZER 1986

Miklós Schweitzer Competition 1986.

Other problems

Problem 1

If (A,<)(A, <) is a partially ordered set, its dimension, dim⁡(A,<)\dim (A, <) , is the least cardinal κ\kappa such that there exist κ\kappa total orderings {<α ⁣:α<κ}\{ <_{\alpha} \colon \alpha < \kappa \} on AA with <=∩α<κ<α<=\cap_{\alpha < \kappa} <_\alpha . Show that if dim⁡(A,<)>ℵ0\dim (A, <)>\aleph_0 , then there exist disjoint A0,A1⊆AA_0, A_1\subseteq A with dim⁡(A0,<)\dim (A_0, <) , dim⁡(A1,<)>ℵ0\dim (A_1, <)>\aleph_0 . [D. Kelly, A. Hajnal, B. Weiss]

Problem 2

Show that if k≤n2k\leq \frac n2 and F\mathcal F is a family k×kk\times k submatrices of an n×nn\times n matrix such that any two intersect then ∣F∣≤(n−1k−1)2|\mathcal F|\leq \binom{n-1}{k-1}^2 [Gy. Katona]

Problem 3

(a) Prove that for every natural number kk , there are positive integers a1<a2<…<aka_1<a_2<\ldots <a_k such that ai−aja_i-a_j divides aia_i for all 1≤i,j≤k,i≠j1\leq i, j\leq k, i\neq j .

(b) Show that there is an absolute constant C>0C>0 such that a1>kCka_1>k^{Ck} for every sequence a1,…,aka_1,\ldots, a_k of numbers that satisfy the above divisibility condition. [A. Balogh, I. Z. Ruzsa]

Problem 4

Determine all real numbers xx for which the following statement is true: the field C\mathbb C of complex numbers contains a proper subfield FF such that adjoining xx to FF we get C\mathbb C . [M. Laczkovich]

Problem 5

Prove that existence of a constant cc with the following property: for every composite integer nn , there exists a group whose order is divisible by nn and is less than ncn^c , and that contains no element of order nn . [P. P. Palfy]

Problem 6

Let UU denote the set {f∈C[0,1] ⁣:∣f(x)∣≤1for allx∈[0,1]}\{ f\in C[0, 1] \colon |f(x)|\leq 1 \mathrm{for}\,\mathrm{all} x\in [0, 1]\} . Prove that there is no topology on C[0,1]C[0, 1] that, together with the linear structure of C[0,1]C[0,1] , makes C[0,1]C[0,1] into a topological vector space in which the set UU is compact. (Assume that topological vector spaces are Hausdorff) [V. Totik]

Problem 7

Prove that the series ∑pcpf(px)\sum_p c_p f(px) , where the summation is over all primes, unconditionally converges in L2[0,1]L^2[0,1] for every 11 -periodic function ff whose restriction to [0,1][0,1] is in L2[0,1]L^2[0,1] if and only if ∑p∣cp∣<∞\sum_p |c_p|<\infty . (Unconditional convergence means convergence for all rearrangements.) [G. Halasz]

Problem 8

Let a0=0a_0=0 , a1,…,aka_1, \ldots, a_k and b1,…,bkb_1, \ldots, b_k be arbitrary real numbers. (i) Show that for all sufficiently large nn there exist polynomials pnp_n of degree at most nn for which pn(i)(−1)=ai,    pn(i)(1)=bi,    i=0,1,…,kp_n^{(i)} (-1)=a_i,\,\,\,\, p_n^{(i)} (1)=b_i,\,\,\,\, i=0, 1, \ldots, k and max⁡∣x∣≤1∣pn(x)∣≤cn2         (∗)\max_{|x|\leq 1} |p_n (x)|\leq \frac{c}{n^2}\,\,\,\,\,\,\,\,\, (*) where the constant cc depends only on the numbers ai,bia_i, b_i .

(ii) Prove that, in general, (*) cannot be replaced by the relation lim⁡n→∞n2⋅max⁡∣x∣≤1∣pn(x)∣=0\lim_{n\to\infty} n^2\cdot \max_{|x|\leq 1} |p_n (x)| = 0 [J. Szabados]

Problem 9

Consider a latticelike packing of translates of a convex region KK . Let tt be the area of the fundamental parallelogram of the lattice defining the packing, and let tmin⁡(K)t_{\min} (K) denote the minimal value of tt taken for all latticelike packings. Is there a natural number NN such that for any n>Nn>N and for any KK different from a parallelogram, ntmin⁡(K)nt_{\min} (K) is smaller that the area of any convex domain in which nn translates to KK can be placed without overlapping? (By a latticelike packing of KK we mean a set of nonoverlapping translates of KK obtained from KK by translations with all vectors of a lattice.) [G. and L. Fejes-Toth]

Problem 10

Let X1,X2X_1, X_2 be independent, identically distributed random variables such that Xi≥0X_i\geq 0 for all ii . Let EXi=m\mathrm EX_i=m , Var(Xi)=σ2<∞\mathrm{Var} (X_i)=\sigma ^2<\infty . Show that, for all 0<α≤10<\alpha\leq 1 lim⁡n→∞n Var([X1+…+Xnn]α)=α2σ2m2(1−α)\lim_{n\to\infty} n\,\mathrm{Var} \left( \left[ \frac{X_1+\ldots +X_n}{n}\right] ^\alpha\right)=\frac{\alpha ^ 2 \sigma ^ 2}{m^{2(1-\alpha)}} [Gy. Michaletzki]