首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 35 毫秒
1.
Let = (1,...,d) be a vector with positive components and let D be the corresponding mixed derivative (of order j with respect to the jth variable). In the case where d > 1 and 0 < k < r are arbitrary, we prove that
and
for all Moreover, if is the least possible value of the exponent in this inequality, then
Deceased.Translated from Ukrainskyi Matematychnyi Zhurnal, Vol. 56, No. 5, pp. 579–594, May, 2004.  相似文献   

2.
We obtain new unimprovable Kolmogorov-type inequalities for differentiable periodic functions. In particular, we prove that, for r = 2, k = 1 or r = 3, k = 1, 2 and arbitrary q, p [1, ], the following unimprovable inequality holds for functions :
where and r is the perfect Euler spline of order r.  相似文献   

3.
For 2-periodic functions and arbitrary q [1, ] and p (0, ], we obtain the new exact Kolmogorov-type inequality which takes into account the number of changes in the sign of the derivatives (x (k)) over the period. Here, = (rk + 1/q)/(r + 1/p), r is the Euler perfect spline of degree r, and . The inequality indicated turns into the equality for functions of the form x(t) = a r (nt + b), a, b R, n N. We also obtain an analog of this inequality in the case where k = 0 and q = and prove new exact Bernstein-type inequalities for trigonometric polynomials and splines.  相似文献   

4.
We obtain the new exact Kolmogorov-type inequality
for 2-periodic functions and any k, r N, k < r. We present applications of this inequality to problems of approximation of one class of functions by another class and estimates of K-functional type.  相似文献   

5.
We extend the results for 2-D Boussinesq equations from ℝ2 to a bounded domain Ω. First, as for the existence of weak solutions, we transform Boussinesq equations to a nonlinear evolution equation U t + A(t, U) = 0. In stead of using the methods of fundamental solutions in the case of entire ℝ2, we study the qualities of F(u, υ) = (u · ▽)υ to get some useful estimates for A(t, U), which helps us to conclude the local-in-time existence and uniqueness of solutions. Second, as for blow-up criterions, we use energy methods, Sobolev inequalities and Gronwall inequality to control and by and . Furthermore, can control by using vorticity transportation equations. At last, can control . Thus, we can find a blow-up criterion in the form of .   相似文献   

6.
Let be the space of 2-periodic functions whose (r – 1)th-order derivative is absolutely continuous on any segment and rth-order derivative belongs to L p, S 2n,m is the space of 2-periodic splines of order m of minimal defect over the uniform partition . In this paper, we construct linear operators such that
where
To construct the operators X n,r,m, we use the same idea as in the polynomial case, i.e., the interpolation of Bernoulli kernels. As is proved, the operators X n,r,m converge to polynomial Akhiezer–Krein–Favard operators as . Bibliography: 10 titles.  相似文献   

7.
  We obtain a new sharp inequality for the local norms of functions x ∈ L ∞, ∞ r (R), namely,
where φ r is the perfect Euler spline, on the segment [a, b] of monotonicity of x for q ≥ 1 and for arbitrary q > 0 in the case where r = 2 or r = 3. As a corollary, we prove the well-known Ligun inequality for periodic functions x ∈ L r , namely,
for q ∈ [0, 1) in the case where r = 2 or r = 3. Translated from Ukrains’kyi Matematychnyi Zhurnal, Vol. 60, No. 10, pp. 1338–1349, October, 2008.  相似文献   

8.
Estimates for deviations are established for a large class of linear methods of approximation of periodic functions by linear combinations of moduli of continuity of different orders. These estimates are sharp in the sense of constants in the uniform and integral metrics. In particular, the following assertion concerning approximation by splines is proved: Suppose that is odd, . Then
moreover, for it is impossible to decrease the constants on . Here, are some explicitly constructed constants, is the modulus of continuity of order r for the function f, and are explicitly constructed linear operators with the values in the space of periodic splines of degree of minimal defect with 2n equidistant interpolation points. This assertion implies the sharp Jackson-type inequality
. Bibliography: 17 titles.  相似文献   

9.
We investigate the relationship between the constants K(R) and K(T), where is the exact constant in the Kolmogorov inequality, R is the real axis, T is a unit circle,
is the set of functions x L p(G) such that x (r) L s(G), q, p, s [1, ], k, r N, k < r, We prove that if
thenK(R) = K(T),but if
thenK(R) K(T); moreover, the last inequality can be an equality as well as a strict inequality. As a corollary, we obtain new exact Kolmogorov-type inequalities on the real axis.  相似文献   

10.
For f L n (T d ) and , the modulus of smoothness
is shown to be equivalent to
where T n is the best trigonometric polynomial approximant of degree n to f in L p and is the Laplacian. The above result is shown to be incorrect for 0 < p .  相似文献   

11.
Let
be the Fejér kernel, C be the space of contiuous 2π-periodic functions f with the norm , let
be the Jackson polynomials of the function f, and let
be the Fejér sums of f. The paper presents upper bounds for certain quantities like
which are exact in order for every function fC. Special attention is paid to the constants occurring in the inequalities obtained. Bibliography: 14 titles. Translated from Zapiski Nauchnykh Seminarov POMI, Vol. 357, 2008, pp. 90–114.  相似文献   

12.
Uniform Approximation of Nonperiodic Functions Defined on the Entire Axis   总被引:1,自引:1,他引:0  
Using the following notation: C is the space of continuous bounded functions f equipped with the norm , V is the set of functions f such that , the set E consists of fCV and possesses the following property:
is summable on each finite interval, we establish some assertions similar to the following theorem: Let 0$$ " align="middle" border="0"> ,
Then for fV the series
uniformly converges with respect to and the following equality holds:
This theorem develops some results obtained by Zubov relative to the approximation of probability distributions. Bibliography: 4 titles.  相似文献   

13.
Summary Letf n (p) be a recursive kernel estimate off (p) thepth order derivative of the probability density functionf, based on a random sample of sizen. In this paper, we provide bounds for the moments of and show that the rate of almost sure convergence of to zero isO(n −α), α<(r−p)/(2r+1), iff (r),r>p≧0, is a continuousL 2(−∞, ∞) function. Similar rate-factor is also obtained for the almost sure convergence of to zero under different conditions onf. This work was supported in part by the Research Foundation of SUNY.  相似文献   

14.
For Banach space operators T satisfying the Tadmor-Ritt condition a band limited H calculus is established, where and a is at most of the order C(T)5. It follows that such a T allows a bounded Besov algebra B∞ 10 functional calculus, These estimates are sharp in a convenient sense. Relevant embedding theorems for B∞ 10 are derived. Received: 25 October 2004; revised: 31 January 2005  相似文献   

15.
Let and . We are interested in the lower bounds of the integral:
where h > 0 and . Using the lower bounds for these integrals we obtain in particular for the so-called Fejér operator of the following asymptotic expression
which essentially improves the results concerning the approximation behavior of this operator. Received: 10 January 2006  相似文献   

16.
Suppose thatА is a nonnegative self-adjoint extension to { } of the formal differential operator−Δu+q(x)u with potentialq(x) satisfying the condition {
} or the condition {
} in which the nonnegative function itχ(r) is such that { }. For each α∈(0, 2], we establish an estimate of the generalized Fourier transforms of an arbitrary function { } of the form {
} If, in addition, { }, then, along with this estimate, a similar lower bound is established. Translated fromMatematicheskie Zametki, Vol. 65, No. 4, pp. 542–551, April, 1999.  相似文献   

17.
Let be the Jacobi polynomials and let C[a,b] be the space of continuous functions on [a,b] with the uniform norm. In this paper, we study sequences of Lebesgue constants, i.e., of the norms of linear operators generated by a multiplier matrix defined by the following relations:
and
In the case || = || = 1/2, we prove the following statements for the Jacobi polynomials (these statements are similar to known results for the trigonometrical system). Consider the cases
and
Under some conditions on a function , the values and equal
and
In addition, we show that for the Fourier–Legendre summation methods ( = = 0) generated by the multiplier function , the limit and supremum of the sequence of Lebesgue constants may differ. Bibliography: 11 titles.  相似文献   

18.
19.
Let be a nondecreasing sequence of positive numbers and let l 1,α be the space of real sequences for which . We associate every sequence ξ from l 1,α with a sequence , where ϕ(·) is a permutation of the natural series such that , j ∈ ℕ. If p is a bounded seminorm on l 1,α and , then
Using this equality, we obtain several known statements. __________ Translated from Ukrains'kyi Matematychnyi Zhurnal, Vol. 57, No. 7, pp. 1002–1006, July, 2005.  相似文献   

20.
We consider the Cauchy problem for the nonlinear dissipative evolution system with ellipticity on one dimensional space
with S. Q. Tang and H. Zhao [4] have considered the problem and obtained the optimal decay property for suitably small data. In this paper we derive the asymptotic profile using the Gauss kernel G(t, x), which shows the precise behavior of solution as time tends to infinity. In fact, we will show that the asymptotic formula
holds, where D0, β0 are determined by the data. It is the key point to reformulate the system to the nonlinear parabolic one by suitable changing variables. (Received: January 8, 2005)  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号