betaFeedback

WorksMathematics

Numerical range of finite perturbations of a 2-periodic weighted shift

Abstract

In this note, we consider a weighted shift operator of weight sequence which is a finite perturbation of a 2-periodic weighted shift. We characterise the equality of the closure of its numerical range and its essential numerical range. We end up with some applications and comments.

Introduction

For a bounded operator TT on a complex Hilbert space HH, the numerical range of TT is

W(T):={⟨Tx,x⟩:x∈H and ∥x∥=1}⊆C.W(T):=\{\langle Tx,x\rangle:x\in H\text{ and }\lVert x\rVert=1\}\subseteq\mathbb{C}.

The numerical radius of TT is w(T):=sup⁡{∣z∣:z∈W(T)}w(T):=\sup\{|z|:z\in W(T)\}. Cauchy-Schwarz inequality leads to w(T)≤∥T∥w(T)\leq\lVert T\rVert. From the definition we have that W(λT)=λW(T)W(\lambda T)=\lambda W(T) for all λ∈C\lambda\in\mathbb{C}.

It is well known that for every operator TT, the set W(T)W(T) is convex and the following inclusions hold σp(T)⊆W(T)\sigma_{p}(T)\subseteq W(T) and σ(T)⊆W(T)\sigma(T)\subseteq W(T) where σp(T)\sigma_{p}(T) and σ(T)\sigma(T) are respectively the point spectrum (i.e. the set of eigenvalues) and the spectrum of TT. Furthermore, W(T)=W(U∗TU)W(T)=W(U^{*}TU) for every unitary operator UU. TT and U∗TUU^{*}TU are said to be unitary equivalent. One may find all useful material about the numerical range and the numerical radius of an operator in [7].

Throughout the paper, the real part of TT will be denoted by Re⁡(T):=T+T∗2\operatorname{Re}(T):=\frac{T+T^{*}}{2}. We have W(Re⁡(T))=Re⁡(W(T))W(\operatorname{Re}(T))=\operatorname{Re}(W(T)).

A self-adjoint operator T∈L(H)T\in\mathcal{L}(H) is called positive if ⟨Tx,x⟩≥0\langle Tx,x\rangle\geq0; ∀x∈H\forall x\in H. We write A≤BA\leq B to express the fact that B−A≥0B-A\geq0.

The essential numerical range of an operator TT is

We(T):=⋂{W(T+K):K is a compact operator}.W_{e}(T):=\bigcap\{W(T+K):K\text{ is a compact operator}\}.

The essential numerical radius of TT is we(T):=sup⁡{∣z∣:z∈We(T)}w_e(T):=\sup\{|z|:z\in W_e(T)\}. It is clear that We(T)W_e(T) is closed and We(T)⊆W(T)W_e(T)\subseteq W(T). A result of Lancaster states that if W(T)W(T) has no extreme point, We(T)=W(T)W_e(T)=W(T) [8 Corollary 2]. In particular, this holds when W(T)W(T) is open.

Let KK be a closed subspace of HH and A,BA,B bounded operators respectively on HH and KK. BB is said to be a compression of AA, or AA is a dilation of BB, if B=πKAπKB=\pi_K A\pi_K where πK\pi_K is the orthogonal projection from HH onto KK. If BB is a compression of AA, we have W(B)⊆W(A)W(B)\subseteq W(A).

For a bounded sequence α=(αn)n≥0∈ℓ∞\alpha=(\alpha_n)_{n\geq0}\in\ell^\infty, the unilateral weighted shift of weight sequence α\alpha denoted by SαS_\alpha maps a sequence (x0,x1,…)∈ℓ2(x_0,x_1,\ldots)\in\ell^2 to (0,α0x0,α1x1,…)(0,\alpha_0x_0,\alpha_1x_1,\ldots). That defines a bounded operator on the Hilbert space ℓ2\ell^2 and ∥Sα∥=∥α∥∞\|S_\alpha\|=\|\alpha\|_\infty. For a sequence α=(αn)n∈ℓ∞\alpha=(\alpha_n)_{n}\in\ell^\infty, let us denote by ∣α∣:=(∣αn∣)n|\alpha|:=(|\alpha_n|)_{n}. It is well known that SαS_\alpha is unitary equivalent to S∣α∣S_{|\alpha|}. Thus, for all α∈ℓ∞\alpha\in\ell^\infty and θ∈[0,2π]\theta\in[0,2\pi], eiθSαe^{i\theta}S_\alpha is unitary equivalent to SαS_\alpha. Hence, the numerical range of a weighted shift is a disk centred at 00 by convexity. Then, to completely describe the numerical range of a weighted shift, we just need to find the radius of the disk and to determine if it is open or closed. For example, the classical shift SS (the weighted shift with weight sequence α=(1,1,…)\alpha=(1,1,\ldots)) has the open unit disk D\mathbb{D} as numerical range.

Since SαS_\alpha is unitary equivalent to S∣α∣S_{|\alpha|} we will only consider non-negative weight sequences in the following. If there is a kk such that αk=0\alpha_k=0, then Sα=S1⊕S2S_\alpha=S_1\oplus S_2 where S1S_1 and S2S_2 are (maybe finite) weighted shifts. Since the numerical range of a direct sum is the convex hull of the union of the numerical ranges, we can restrict ourselves to positive weight sequences.

In the following, we will denote by D(0,r)D(0,r) and D‾(0,r)\overline{D}(0,r) respectively the open and closed disk of radius r≥0r\geq0 centred at 00.

The numerical range of weighted shifts with periodic weight sequence was first studied by Ridge in [10]. He proved that a weighted shift with periodic weight sequence (a1,…,an,a1,…,an,a1,…)(a_1,\ldots,a_n,a_1,\ldots,a_n,a_1,\ldots) has the same numerical radius as the weighted cyclic matrix

(0ana10⋱⋱an−10)\begin{pmatrix} 0 & & & a_n \\ a_1 & 0 & & \\ & \ddots& \ddots& \\ & & a_{n-1} & 0 \end{pmatrix}

Thus, for n=2n=2, we know that w(Sα)=(a+b)/2w(S_\alpha)=(a+b)/2 if α=(a,b,a,b,a,…)\alpha=(a,b,a,b,a,\ldots). Nevertheless, we do not know the numerical radius of periodic shifts with period bigger than 3. Indeed, in dimension bigger than 3, the numerical radius of weighted cyclic matrices is unknown. This is why we will focus on 2-periodic weighted shifts in the sequel.

Ridge conjectured that the numerical range of periodic weighted shifts are open. That was settled in positive by Stout in [11] (see also [12 Theorem 4] and [15 Proposition 4.2]). Thus, we can claim that for every periodic sequence α\alpha, We(Sα)=W(Sα)‾W_e(S_\alpha)=\overline{W(S_\alpha)} by Lancaster’s Theorem cited above. Fix nn an even number, a1,…,an>0a_1,\ldots,a_n>0 and a,b>0a,b>0 and define α:=(a1,…,an,a,b,a,b,…)\alpha:=(a_1,\ldots,a_n,a,b,a,b,\ldots). SαS_\alpha is a finite (thus compact) perturbation of the 2-periodic shift with weight sequence (a,b,a,b,…)(a,b,a,b,\ldots).

According to the preceding, the situation is as follows:

  1. W(Sα)W(S_\alpha) is a disk centred at 00 with radius (a+b)/2=we(Sα)(a+b)/2=w_e(S_\alpha);

  1. or W(Sα)W(S_\alpha) is a closed disk centred at 00 with radius r>(a+b)/2r>(a+b)/2.

Theorem 4.3 of [15] ensures that in the first case, W(Sα)W(S_\alpha) is open. We will give another proof of this result at the end of the article in our context of 2-periodic shifts. Our goal in this article is to give a characterisation of when the first case appears. That is our Theorem 2.5. We will follow the philosophy of the article of Wang and Wu [15], and will not try to express the numerical radius in the second case.

This question has already been studied in the past. In 1967, Berger and Stampfli treated the case n=1n=1 and a=b=1a=b=1 in [1]. If t>0t>0 and α=(t,1,1,…)\alpha=(t,1,1,\ldots), W(Sα)=DW(S_\alpha)=\mathbb{D} if t≤2t\le\sqrt{2} and equals D‾(0,t22t2−1)\overline{\mathbb{D}}\left(0,\frac{t^2}{2\sqrt{t^2-1}}\right) if t>2t>\sqrt{2}. The proof uses Hardy space techniques. The fact that W(Sα)=DW(S_\alpha)=\mathbb{D} is shown to be equivalent for the polynomial (t2−1)z2−1(t^2-1)z^2-1 to have no root in (0,1)(0,1).

This approach has been followed in [15 Corollary 4.8] (see also [4]) and [14] respectively for the weight sequence α=(1,h,1,1,…)\alpha=(1,h,1,1,\ldots) and α=(a1,a2,1,1,…)\alpha=(a_1,a_2,1,1,\ldots). We have w(Sα)=1w(S_\alpha)=1 if and only if h≤6/2h\le\sqrt{6}/2 and 4−a12−2a22≥04-a_1^2-2a_2^2\ge0 respectively. Otherwise, the numerical radius of SαS_\alpha is provided.

The case of an arbitrary finite perturbation α=(a1,…,an,1,1,…)\alpha=(a_1,\ldots,a_n,1,1,\ldots) was first considered in [13]. The authors used the same method as above and exhibit a polynomial GαG_\alpha such that w(Sα)>1w(S_\alpha)>1 if and only if GαG_\alpha has a real root in (0,1)(0,1) (Theorem 3.3 of [13]). The numerical radius is no longer explicit there, but is expressed in terms of the smallest root of GαG_\alpha in (0,1)(0,1).

References

  1. [1]CA Berger and JG Stampfli. Mapping theorems for the numerical range. American Journal of Mathematics, 89(4):1047–1055, 1967.
  2. [2]Bikshan Chakraborty, Sarita Ojha, and Riddhick Birbonshi. Numerical radii of weighted shift operators using determinantal polynomials. Oper. Matrices, 16(4):1155–1174, 2022.DOI
  3. [3]Bikshan Chakraborty, Sarita Ojha, and Riddhick Birbonshi. On the numerical range of some weighted shift operators. Linear Algebra and its Applications, 640:179–190, 2022.DOI
  4. [4]Mao-Ting Chien and Hue-An Sheu. The numerical radii of weighted shift matrices and operators. Oper. Matrices, 7(1):197–204, 2013.DOI
  5. [5]Chi-Tung Chang and Kuo-Zhong Wang. Maximizing numerical radii of weighted shifts under weight permutations. Journal of Mathematical Analysis and Applications, 394(2):592–602, 2012.DOI
  6. [6]G Eckstein and A Racz. Weighted shifts of class 𝒞ρ. ACTA SCIENTIARUM MATHEMATICARUM, 35:187–194, 1973.
  7. [7]Hwa-Long Gau and Pei Yuan Wu. Numerical ranges of Hilbert space operators, volume 179. Cambridge University Press, 2021.DOI
  8. [8]John S Lancaster. The boundary of the numerical range. Proceedings of the American Mathematical Society, pages 393–398, 1975.DOI
  9. [9]Rubén A Martínez-Avendaño and Peter Rosenthal. An introduction to operators on the Hardy-Hilbert space. Springer, 2007.
  10. [10]William C Ridge. Numerical range of a weighted shift with periodic weights. In Proc. Amer. Math. Soc, volume 55, pages 107–110, 1976.DOI
  11. [11]Quentin F Stout. The numerical range of a weighted shift. Proceedings of the American Mathematical Society, 88(3):495–502, 1983.DOI
  12. [12]Tin-Yau Tam. On a conjecture of ridge. Proceedings of the American Mathematical Society, 125(12):3581–3592, 1997.DOI
  13. [13]Batzorig Undrakh, Hiroshi Nakazato, Adiyasuren Vandanjav, and Mao-Ting Chien. The numerical radius of a weighted shift operator. The Electronic Journal of Linear Algebra, 30:944–963, 2015.DOI
  14. [14]Adiyasuren Vandanjav and Batzorig Undrakh. On the numerical range of some weighted shift matrices and operators. Linear Algebra and its Applications, 449:76–88, 2014.DOI
  15. [15]Kuo-Zhong Wang and Pei Yuan Wu. Numerical ranges of weighted shifts. Journal of Mathematical Analysis and Applications, 381(2):897–909, 2011.DOI

Paper details

vlnc:W_9m5jy8hv
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents