Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy

Solutions of the discrete Painlevé II hierarchy are shown to be in relation to a family of Toeplitz determinants describing certain quantities in multicritical random partition models, for which the limiting behavior has been recently considered in the literature. Our proof is based on the Riemann-H...

Повний опис

Збережено в:
Бібліографічні деталі
Опубліковано в: :Symmetry, Integrability and Geometry: Methods and Applications
Дата:2023
Автори: Chouteau, Thomas, Tarricone, Sofia
Формат: Стаття
Мова:Англійська
Опубліковано: Інститут математики НАН України 2023
Онлайн доступ:https://nasplib.isofts.kiev.ua/handle/123456789/211913
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Цитувати:Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy. Thomas Chouteau and Sofia Tarricone. SIGMA 19 (2023), 030, 30 pages

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine
_version_ 1859868056064360448
author Chouteau, Thomas
Tarricone, Sofia
author_facet Chouteau, Thomas
Tarricone, Sofia
citation_txt Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy. Thomas Chouteau and Sofia Tarricone. SIGMA 19 (2023), 030, 30 pages
collection DSpace DC
container_title Symmetry, Integrability and Geometry: Methods and Applications
description Solutions of the discrete Painlevé II hierarchy are shown to be in relation to a family of Toeplitz determinants describing certain quantities in multicritical random partition models, for which the limiting behavior has been recently considered in the literature. Our proof is based on the Riemann-Hilbert approach for the orthogonal polynomials on the unit circle related to the Toeplitz determinants of interest. This technique allows us to construct a new Lax pair for the discrete Painlevé II hierarchy that is then mapped to the one introduced by Cresswell and Joshi.
first_indexed 2026-03-17T00:46:36Z
format Article
fulltext Symmetry, Integrability and Geometry: Methods and Applications SIGMA 19 (2023), 030, 30 pages Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy Thomas CHOUTEAU a and Sofia TARRICONE b a) Université d’Angers, CNRS, LAREMA, SFR MATHSTIC, F-49000 Angers, France E-mail: thomas.chouteau@univ-angers.fr b) Institut de Physique Théorique, Université Paris-Saclay, CEA, CNRS, F-91191 Gif-sur-Yvette, France E-mail: sofia.tarricone@ipht.fr URL: https://starricone.netlify.app/ Received December 22, 2022, in final form May 16, 2023; Published online May 28, 2023 https://doi.org/10.3842/SIGMA.2023.030 Abstract. Solutions of the discrete Painlevé II hierarchy are shown to be in relation with a family of Toeplitz determinants describing certain quantities in multicritical random parti- tions models, for which the limiting behavior has been recently considered in the literature. Our proof is based on the Riemann–Hilbert approach for the orthogonal polynomials on the unit circle related to the Toeplitz determinants of interest. This technique allows us to construct a new Lax pair for the discrete Painlevé II hierarchy that is then mapped to the one introduced by Cresswell and Joshi. Key words: discrete Painlevé equations; orthogonal polynomials; Riemann–Hilbert prob- lems; Toeplitz determinants 2020 Mathematics Subject Classification: 33E17; 33C47; 35Q15 1 Introduction Let us consider the symbol φ(z) = ew(z) with w(z) := v(z) + v ( z−1 ) and v(z) := N∑ j=1 θj j zj , (1.1) for θj being real constants and natural N ≥ 1. The n-th Toeplitz matrix associated to this symbol and denoted by Tn(φ) is a square (n+1)-dimensional matrix which entries are given by Tn(φ)i,j := φi−j , i, j = 0, . . . , n. (1.2) Here for every k ∈ Z, φk is the k-th Fourier coefficient of φ(z), namely φk = ∫ π −π e−ikβφ ( eiβ )dβ 2π , so that ∑ k∈Z φkz k = φ(z). Notice that, even though it is not emphasized in our notation, the functions φk and thus the Toeplitz matrix Tn(φ) explicitly depend on the natural parameter N which enters in the definition of v(z) in equation (1.1). This paper is a contribution to the Special Issue on Evolution Equations, Exactly Solvable Mod- els and Random Matrices in honor of Alexander Its’ 70th birthday. The full collection is available at https://www.emis.de/journals/SIGMA/Its.html mailto:thomas.chouteau@univ-angers.fr mailto:sofia.tarricone@ipht.fr https://starricone.netlify.app/ https://doi.org/10.3842/SIGMA.2023.030 https://www.emis.de/journals/SIGMA/Its.html 2 T. Chouteau and S. Tarricone In the present work, it is indeed the dependence on this parameter N that we want to study. In particular, we show that the Toeplitz determinants associated to Tn(φ), naturally defined as DN n := Dn = det(Tn(φ)), (1.3) are related to some solutions of a discrete version of the Painlevé II hierarchy, indexed over the parameter N (the dependence on N is dropped in the rest of the paper). Our interest in these Toeplitz determinants comes from their appearance in the recent paper [5]. The authors there consider some probability measures on the set of integer partitions called multicritical Schur measures, which are a particular case of Schur measures introduced by Okounkov in [23]. They are generalizations of the classical Poissonized Plancherel measure and they are defined as P({λ}) = Z−1sλ[θ1, . . . , θN ]2, with Z = exp ( N∑ i=1 θ2i i ) . (1.4) Here sλ[θ1, . . . , θN ] denotes a Schur symmetric function indexed by a partition λ that can be expressed as sλ[θ1, . . . , θN ] = det i,j hλi−i+j [θ1, . . . , θN ], where ∑ k≥0 hkz k = exp (∑N i=1 θi i z i ) . In [5], the authors first used the term multicritical to un- derline that they obtained a different limiting edge behavior for these Schur measures compared to the classical case of the Poissonized Plancherel measure (N = 1) which is characterised by the Tracy–Widom GUE distribution. For more details, we remind to their Theorem 1 or our discussion in the paragraph “Continuous limit” below, for instance see equation (1.23) where the higher order Tracy–Widom distributions appear. In this setting, denoting by λ = (λ1 ≥ λ2 ≥ · · · ≥ 0) a generic integer partition and by λ′ = (λ′ 1 ≥ λ′ 2 ≥ · · · ≥ 0) its conjugate partition (namely such that λ′ j = |i : λi ≥ j|), major quantities of interest of the model are, for any given n ∈ N, rn := P(λ1 ≤ n) and qn := P(λ′ 1 ≤ n), (1.5) that are often called discrete gap probabilities as random partitions have a natural interpretation in terms of random configuration of points on the set of semi-integers. Indeed, associating the set {λi − i + 1/2} ⊂ Z + 1 2 to a partition λ (see [23]), rn and qn can be expressed in terms of a Fredholm determinant of a discrete kernel which corresponds to the gap probability in the determinantal point process defined through the same kernel. According to Geronimo–Case/Borodin–Okounkov formula [7], there is a relation between this Fredholm determinant and the Toeplitz determinant Dn and this implies that rn and qn (up to a constant factor) are Toeplitz determinants. It leads to (for instance [5, Propositions 6 and 7]): qn = e− ∑N j=1 θ 2 j /jDn−1. (1.6) For rn instead, one should define θ̃i = (−1)i−1θi and by taking w̃(z) = ṽ(z) + ṽ ( z−1 ) , where ṽ(z) is nothing than v(z) with θi replaced by θ̃i as given above, the Toeplitz determinant D̃n associated to the symbol φ̃(z) = ew̃(z) would give the analogue formula rn = e− ∑N j=1 θ̃ 2 j /jD̃n−1. Notice that in the simplest case, when N = 1, the quantities rn and qn coincide. Moreover, thanks to Schensted’s theorem [27], they are also equal to the discrete probability distribution Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 3 function of the length of the longest increasing subsequence of random permutations of size m, with m distributed as a Poisson random variable. In the case N = 1, the relation of these quantities with the theory of discrete Painlevé equations was shown two decades ago independently and through very different methods by Borodin [6], Baik [2], Adler and van Moerbeke [1] and Forrester and Witte [16].1 In particular, they all proved that for every n ≥ 1, the following chain of equalities holds DnDn−2 D2 n−1 = qn+1qn−1 q2n = rn+1rn−1 r2n = 1− x2n, (1.7) where xn solves the second order nonlinear difference equation θ1(xn+1 + xn−1) ( 1− x2n ) + nxn = 0, (1.8) with certain initial conditions. Equation (1.8) is a particular case of the so called discrete Painlevé II equation [26], a discrete analogue of the classical second order ODE known as the Painlevé II equation [24]. This means that performing some continuous limit of equation (1.8) one gets back the Painlevé II equation. The Painlevé II equations, discrete and continuous ones, depend in general on an additional constant term α ∈ R. In the present work, we consider the discrete Painlevé II equation and its hierarchy in the homogeneous case where α = 0. Its continuous limit will correspond as well to the case α = 0. Remark 1.1. The homogeneous Painlevé II equation admits a famous solution [17], called the Hastings–McLeod solution, found by requiring a specific boundary condition at ∞. In parallel, one might wonder what is the large n behavior of the solution xn of the discrete Painlevé II equation (1.8). Its behavior is expressed in terms of the Bessel functions. First, this is suggested by the following heuristic arguments. Because of the definition of rn (1.5), as n → ∞, rn tends to one and according to the equation (1.7), xn tends to zero. Then for large n, the nonlinear term in equation (1.8) is small compared to the linear ones and the equation (1.8) reduces to the equation θ1 (xn+1 + xn−1) + nxn = 0, which indeed admits J−n(2θ1), the Bessel function of the first kind of order −n, as a solution. The claim is confirmed by a result of the recent work [9]. The authors there studied the finite temperature deformation for the discrete Bessel point process. The Fredholm determinant of the finite temperature discrete Bessel kernel they studied depends on a function σ. In the case when σ = 1Z′ + (the characteristic function of the set of positive half integers), the Fredholm determinant is then equal to rn. Then from [9, equations (1.33) and (1.36) of Theorem III] together with equation (1.7), one can deduce that for n large x2n ∼ Jn(2θ1) 2 and, because of the previous discussion, one can conclude xn ∼ J−n(2θ1) = (−1)nJn(2θ1), see also Figure 1. For N > 1, Adler and van Moerbeke presented in [1], a generalization of equation (1.7) by proving that xn satisfies some recurrence relation written in terms of the Toeplitz lattice Lax matrices. The main result of our work is a recurrence relation for xn defined via a N -times iterating discrete operator which establishes the link with the discrete Painlevé II hierarchy [11]. The precise result is stated as below. 1They obtained an analogue of equation (1.7) for Toeplitz determinant associated to symbols which are not necessarily positive or even real valued. 4 T. Chouteau and S. Tarricone 0 10 20 30 40 50 n 1.00 0.75 0.50 0.25 0.00 0.25 0.50 0.75 1.00 xn ( 1)nJn(2 1) Figure 1. For N = 1, the graphs of xn and (−1)nJn(2θ1) in function of n for θ1 = 3. Theorem 1.2. For any fixed N ≥ 1, for the Toeplitz determinants Dn (1.3), n ≥ 1 associated to the symbol φ(z) (1.1), we have DnDn−2 D2 n−1 = 1− x2n, (1.9) where xn solves the 2N order nonlinear difference equation nxn + ( −vn − vnPermn + 2xn∆ −1(xn − (∆ + I)xnPermn) ) LN (0) = 0, (1.10) where L is a discrete recursion operator defined as L(un) := ( xn+1 ( 2∆−1 + I ) ((∆ + I)xnPermn − xn) + vn+1(∆ + I)− xnxn+1 ) un. (1.11) Here vn := 1− x2n, ∆ denotes the difference operator ∆: un → un+1 − un and Permn is the transformation of the space C [ (xj)j∈[[0,2n]] ] acting by permuting indices in the following way: Permn : C [ (xj)j∈[[0,2n]] ] −→ C [ (xj)j∈[[0,2n]] ] , P ( (xn+j)−n⩽j⩽n ) 7−→ P ( (xn−j)−n⩽j⩽n ) . (1.12) Remark 1.3. According to equation (1.10) and the definition of the operator L (1.11), we need to perform discrete integrations to compute the N -th equation of the discrete Painlevé II hierarchy. It is always possible to accomplish this discrete integration. The operator ∆−1, inverse of the difference operator ∆, is applied to (∆ + I)xnPermn − xn and it is possible to write this operator as a derivative. Indeed, (∆ + I)xnPermn − xn = ∆xnPermn + (Permn − I)xn. The first term on the right hand side is a derivative and because of the definition of Permn, the second term can be expressed as a derivative. Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 5 Equation (1.10), together with the definition of the recursion operator L in (1.11), of the quantity vn and of the transformation Permn in (1.12) is indeed the N -th member of the discrete Painlevé II hierarchy. The first equations of the hierarchy read as N = 1: nxn + θ1(xn+1 + xn−1) ( 1− x2n ) = 0, (1.13) N = 2: nxn + θ1 ( 1− x2n ) (xn+1 + xn−1) + θ2 ( 1− x2n ) × ( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 ) − xn(xn+1 + xn−1) 2 ) = 0, (1.14) N = 3: nxn + θ1 ( 1− x2n ) (xn+1 + xn−1) + θ2 ( 1− x2n )( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 ) − xn(xn+1 + xn−1) 2 ) + θ3 ( 1− x2n )( x2n(xn+1 + xn−1) 3 + xn+3 ( 1− x2n+2 )( 1− x2n+1 ) + xn−3 ( 1− x2n−2 )( 1− x2n−1 )) + θ3 ( 1− x2n )( −2xn(xn+1 + xn−1) ( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 )) − xn−1x 2 n−2 ( 1− x2n−1 )) + θ3 ( 1− x2n )( −xn+1x 2 n+2 ( 1− x2n+1 ) − xn+1xn−1(xn+1 + xn−1) ) = 0 (1.15) with the first one coinciding with the discrete Painlevé II equation (1.8). Computations with the operator (1.11) introduced in Theorem 1.2 for N = 1 and 2 are done in Example 3.11. Remark 1.4. The same heuristic argument used in Remark 1.1 applies also when N > 1 (since xn still tends to zero as n → ∞), thus suggesting that the N -th equation of the discrete Painlevé II hierarchy reduces to a linear discrete equation for large n. For N = 2 and 3, the reduced equations are N = 2: nxn + θ1(xn+1 + xn−1) + θ2(xn+2 + xn−2) = 0, N = 3: nxn + θ1(xn+1 + xn−1) + θ2(xn+2 + xn−2) + θ3(xn+3 + xn−3) = 0. Similar recurrence relations appeared in [12] for the multivariable generalized Bessel functions (GBFs). These generalized Bessel functions were discussed in [21, 23] in the context of Schur measures for random partitions and generalizations of the previous recurrence equations were introduced (in particular, see in [21, equation (3.2b)]). We denote by J (N) n (ξ1, . . . , ξN ) a N - variable GBFs of order n. In [12], J (N) n (ξ1, . . . , ξN ) is defined as a discrete convolution product of N Bessel functions. In particular, if j (k) n (ξ) is the n-th Fourier coefficient of the function β → e2iξ sin(kβ) then J (N) n (ξ1, . . . , ξN ) := j(N) n (ξN ) ∗ j(N−1) n (ξN−1) ∗ · · · ∗ j(1)n (ξ1)(n), where ∗ denotes the discrete convolution. In the case N = 1, the symbol we considered was φ ( eiβ ) = eθ1(e iβ+e−iβ) = e2θ1 cos(β) and the large n asymptotic behavior of xn was given by J−n(2θ1) which is the n-th Fourier coefficient of the function β → eθ1(e iβ−e−iβ) up to a constant (−1)n. For N > 1, the symbol is φN (eiβ) = ∏N k=1 e θk k (eikβ+e−ikβ) = ∏N k=1 e 2 θk k cos(kβ). Then, we conjecture that the large n asymptotic behavior of x (N) n would be given by the n-th Fourier coefficient of β → ∏N k=1 e (−1)k+1θk k (eikβ−e−ikβ) which is precisely J (N) n (ξ1, . . . , ξN ) up to some constant and proper rescaling: x(N) n ∼ (−1)nJ (N) n (( (−1)i 2 i θi ) 1⩽i⩽N ) , see also Figure 2. 6 T. Chouteau and S. Tarricone 0 10 20 30 40 50 n 1.00 0.75 0.50 0.25 0.00 0.25 0.50 0.75 1.00 x(2) n ( 1)nJ(2) n (2 1, 2) 0 10 20 30 40 50 n 1.00 0.75 0.50 0.25 0.00 0.25 0.50 0.75 1.00 x(3) n ( 1)nJ(3) n (2 1, 2, 2/3 3) Figure 2. The graphs of x (N) n and (−1)nJ (N) n ( (θi)1⩽i⩽N ) (for N = 2 on left and N = 3 on the right) in function of n for θ1 = 3, θ2 = 1.2 and θ3 = 2.6. Remark 1.5. Notice that for N = 1, 2 the equations (1.13) and (1.14) coincide with the ones found in [1]. Also notice that in the physics literature, Periwal and Schewitz [25] found similar discrete equations for N = 1, 2 (with different coefficients sign) in the context of unitary matrix models and used their solutions to evaluate the behavior of some typical integrals in the large- dimensional limit passing through the continuous limit of their discrete equations. For N = 1, the discrete Painlevé II equation was also found in [18] as a particular case of the string equation for the full unitary matrix model, i.e., for w(z) = θ1z + θ−1z −1. The dependence in θ±1 of xn (and some other x∗n) was also studied there and it produced some evolution equations related, after some change of variables, to the two-dimensional Toda equations. This would suggest that for the general case N > 1, the dependence of xn on times θ1, . . . , θN would be related to the one-dimensional Toda hierarchy (see also [23]). The first construction of a discrete Painlevé II hierarchy in [11] used the integrability property of the continuous one, in the following sense. It is well known that the classical Painlevé II equa- tion admits an entire hierarchy of higher order analogues. Indeed, this equation can be obtained as a self-similarity reduction of the modified KdV equation. Thus, the higher order members of the Painlevé II hierarchy are but analogue self-similarity reductions of the corresponding higher order members of the modified KdV hierarchy (see, e.g., [14]). In some way, this implies that the Lax representation of the KdV hierarchy in terms of isospectral deformations becomes for the Painlevé II hierarchy a Lax representation in terms of isomonodromic deformations [10]. In [11] then, the discrete Painlevé II hierarchy is defined as the compatibility condition of a sort of “discretization” of the Lax representation of the Painlevé II hierarchy. In particular, they considered the compatibility condition of a linear 2×2 matrix-valued system of the following type: Φn+1(z) = Ln(z)Φn(z), ∂ ∂z Φn(z) = Mn(z)Φn(z), (1.16) where the coefficients Ln(z), Mn(z) are explicit matrix-valued rational function in z, depending on xℓ, ℓ = n+N, . . . , n−N , in some recursive (on N) way. This allows the authors there to com- Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 7 pactly write the N -th discrete Painlevé II equation using some recursion operators. The linear system that we obtain in Proposition 2.11 and that encodes our hierarchy as written in (1.10) is mapped into the one of [11] through an explicit transformation, as shown in Proposition 2.18, thus implying that (1.10) is indeed the same discrete Painlevé II hierarchy. Continuous limit. The aim of this paragraph is to explain heuristically the reason why our result given in Theorem 1.2 can be considered as the discrete analogue of the generalized Tracy– Widom formula for higher order Airy kernels (namely, the result contained in [8, Theorem 1.1], case τi = 0). For N = 1, Borodin in [6] already pointed out that formula (1.7) with (1.8) can be seen as a discrete analogue of the classical Tracy–Widom formula for the GUE Tracy–Widom distribu- tion [28, 29]. In other words, he described how to pass from the left to the right in the picture below: Discrete case DnDn−2 −D2 n−1 D2 n−1 = −x2n with nxn + θ ( 1− x2n ) (xn+1 + xn−1) = 0, Continuous case d2 dt2 log det(1−KAi|(t,+∞)) = −u2(t) with u′′(t) = 2u3(t) + tu(t), u(t) ∼ t→∞ Ai(t), Baik–Deift–Johansson where Ai(t) denotes the classical Airy function and KAi denotes the integral operator acting on L2(R) through the Airy kernel. This connection was achieved by using the scaling limit computed by Baik, Deift and Johansonn in [3] for the distribution of the first part of partitions in the Poissonized Plancherel random partition model (which is recovered in [5, Theorem 1] for N = 1). In some way, as emphasized by Borodin, their result not only assures the existence of a limiting function for the Dn, in this case D(t) = det(1−KAi|(t,+∞)), for a certain continuous variable t. It also encodes already how the discrete function xn, should be rescaled in terms of a differentiable function u(t) to get back, from the recursion relation for Dn, the Tracy–Widom formula. To generalize this result for the case N > 1, we proceed by adapting the method used by Borodin in [6] for N = 1 to the higher order cases, using the scaling proposed in [5]2 for the multicritical case (notice that their n corresponds to our N), instead of the Baik–Deift– Johansson’s one that only holds for N = 1. We recall that Dn is the Toeplitz determinant associated to the symbol φ(z) (1.1) (which depends on θi, i = 1, . . . , N and thus on N). In the following discussion, we write explicitly the dependence on the family of parameters (θi), i = 1, . . . , N of Dn = Dn(θi), xn = xn(θi), rn = rn(θi) and qn = qn(θi). Consider equation (1.9) written in terms of the Toeplitz determi- nants Dn(θi) in this way Dn−2(θi)Dn(θi)−D2 n−1(θi) D2 n−1(θi) = −x2n(θi). (1.17) From the equation (1.6), this previous equation can be expressed in terms of qn(θi) defined as (1.5). It becomes qn−1(θi)qn+1(θi)− q2n(θi) q2n(θi) = −x2n(θi). (1.18) 2Up to the correction of the typo d → d−1 in their statement of Theorem 1. 8 T. Chouteau and S. Tarricone According to [5, Lemma 8], with the change of parameters θ̃i = (−1)i−1θi, we have qn(θi) = rn ( θ̃i ) . Thus equation (1.18) now reads as rn−1 ( θ̃i ) rn+1 ( θ̃i ) − r2n ( θ̃i ) r2n ( θ̃i ) = −x2n(θi). (1.19) Following the scaling limit described in [5, Theorem 1], we define the following scaling for the discrete variable n: n = bθ + tθ 1 2N+1d− 1 2N+1 ⇐⇒ t = (n− bθ)θ− 1 2N+1d 1 2N+1 (1.20) with b, d defined as b = N + 1 N , d = ( 2N N − 1 ) and choose θ̃i (resp. θi) all proportional to θ = θ̃1 = θ1 in the following way: θ̃i = (−1)i−1 (N − 1)!(N + 1)! (N − i)!(N + i)! θ, i = 1, . . . , N, respectively, θi = (N − 1)!(N + 1)! (N − i)!(N + i)! θ, i = 1, . . . , N. (1.21) Now recall the definition of rn ( θ̃i ) (1.5) in function of P = Pθ̃i (see equation (1.4) for the definition of P and the dependence on the family of parameters (θi)). From the previous scaling, it is now possible to express rn ( θ̃i ) in function of t and θ rn ( θ̃i ) = Pθ̃i ( λ1 − bθ (θd−1) 1 2N+1 ⩽ t ) (1.22) and according to [5, Theorem 1], the limiting behavior of the probability distribution function of λ1 in this setting is given by lim θ→+∞ rn ( θ̃i ) = lim θ→+∞ Pθ̃i ( λ1 − bθ( θd−1 ) 1 2N+1 ⩽ t ) = FN (t), FN (t) = det(1−KAi2N+1 |(t,∞)), (1.23) where KAi2N+1 is the integral operator acting with higher order Airy kernel (see, for instance, in [5, equation (2.7)]). As we did for rn ( θ̃i ) in equation (1.22), we express rn+1 ( θ̃i ) and rn−1 ( θ̃i ) in function of t and θ: rn±1 ( θ̃i ) = Pθ̃i ( λ1 − bθ( θd−1 ) 1 2N+1 ⩽ t± ( θd−1 )− 1 2N+1 ) . With this discussion and this scaling for n, (θi) and ( θ̃i ) , we deduce that − lim θ→+∞ x2n(θi)( θd−1 )− 2 2N+1 = lim θ→+∞ rn−1 ( θ̃i ) rn+1 ( θ̃i ) − r2n ( θ̃i )( θd−1 )− 2 2N+1 r2n ( θ̃i ) = d2 dt2 logFN (t), where the first equality comes from equation (1.19) and the second from equation (1.23). Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 9 From now on, we drop the dependence on θi, i = 1, . . . , N in the notation. The previous equation suggests that, in order to be consistent with [8, Theorem 1.1], the discrete function xn appearing in formula (1.17) in the scaling (1.20) for n and (1.21) for (θi) limit should be −x2n ∼ −(θ)− 2 2N+1d 2 2N+1u2(t) with u(t) solution of the N -th equation of the Painlevé II hierarchy. This can be proved directly by computing the scaling limit of the equations of the discrete Painlevé II hierarchy we found for xn in Theorem 1.2. Indeed, for every fixed N , we write xn as xn = (−1)nθ− 1 2N+1d 1 2N+1u(t) (1.24) with u(t) a smooth function of the variable t defined as in equation (1.20). Now recall that xn solves the discrete equation (1.10) of order 2N for every N ≥ 1. The continuous limit of the discrete equations of the hierarchy (1.10), under the definition of xn (1.24) and the scaling of the parameters θi as (1.21), gives the equations of the classical Painlevé II hierarchy. For any fixed N the computation should be done in the same way: consider the N -th discrete equation of the hierarchy (1.10) and replace each θi with the values given in formula (1.21). Then substitute xn with the definition in (1.24) and for θ → +∞ compute the asymptotic expansion of every term xn+K ∝ u ( t +Kθ− 1 2N+1d 1 2N+1 ) , K = −N, . . . , N appearing in the discrete equation. The coefficient of θ−1 resulting after this procedure coincides indeed with the N -th equation of the Painlevé II hierarchy. For N = 1, 2, 3, the computations are explicitly done in the Appendix A. Remark 1.6. It is worthy to mention that in [8], the authors also consider a generalization of the Fredholm determinant FN (t), recalled here in (1.23), depending on additional parameters τi. Those are related to solutions of the general Painlevé II hierarchy, which depends as well on the τi. With the scaling as in [5] for the θi’s, the continuous limit for our discrete equations leads to the Painlevé II hierarchy with τi = 0 for all i. This is consistent with the fact that the limiting behavior in [5], written here in equation (1.23), involves indeed the Fredholm determinant FN (t) corresponding to τi = 0 for all i (the same already appeared in [22]). Methodology and outline. The rest of the work is devoted to prove Theorem 1.2. In order to do so, we introduce the classical Riemann–Hilbert characterization [4] of the family of orthog- onal polynomials on the unit circle (OPUC for brevity) with respect to a measure defined by the symbol φ(z). Classical results from orthogonal polynomials theory allow to achieve almost directly formula (1.17) where xn is defined as the constant term of the n-th monic orthogonal polynomial of the family. The Riemann–Hilbert problem for the OPUC is then used to deduce a linear system of the same type of (1.16) which is proven to be in relation with the Lax pair introduced by Cresswell and Joshi [11] for the discrete Painlevé II hierarchy. This is done in Sec- tion 2. The explicit computation of the Lax pair together with the construction of the recursion operator and the hierarchy for xn as written in (1.10) are done in Section 3. 2 OPUC: the Riemann–Hilbert approach and a discrete Painlevé II Lax pair In this section, we introduce the relevant family of orthogonal polynomials on the unit circle. We recall some of their properties and their Riemann–Hilbert characterization. Afterward we derive a Lax pair associated to the Riemann–Hilbert problem and establish the relation with the Lax pair for discrete Painlevé II hierarchy (1.16) introduced by Cresswell and Joshi [11]. The proofs of the results for orthogonal polynomials stated in here can be found in the classical reference [4]. 10 T. Chouteau and S. Tarricone We denote by S1 the unit circle in C counterclockwise oriented. We consider the following positive measure on S1 (absolutely continuous w.r.t. the Lebesgue measure there): dµ(β) = ew(eiβ) 2π dβ, (2.1) where the function w(z) for any z ∈ C is given as in equation (1.1). The family of orthogonal polynomials on the unit circle (OPUC) w.r.t. the measure (2.1) is defined as the collection of polynomials {pn(z)}n∈N written as pn(z) = κnz n + · · ·+ κ0, κn > 0 (2.2) and such that the following relation holds for any indices k, h∫ π −π pk ( eiβ ) ph ( eiβ )dµ(β) 2π = δk,h. The family of monic orthogonal polynomials {πn(z)} associated to the previous ones is defined in analogue way, so that pn(z) = κnπn(z). 2.1 Toeplitz determinants related to OPUC We recall that φ(z) = ew(z), z ∈ S1 with w(z) defined as in (1.1) and that we defined Dn := det(Tn(φ)) (by conventionD−1 = 1) to be the n-Toeplitz determinant associated to the symbol φ (see equations (1.2) and (1.3)). Because φ(z) is a real nonnegative function, Dn ∈ R>0. Proposition 2.1. If φ(z) is a real nonnegative function, we have that pℓ(z) = 1√ DℓDℓ−1 det  φ0 φ−1 . . . φ−ℓ+1 φ−ℓ φ1 φ0 . . . φ−ℓ+2 φ−ℓ+1 ... ... . . . ... ... φℓ−1 φℓ−2 . . . φ0 φ−1 1 z . . . zℓ−1 zℓ , ℓ ≥ 0. (2.3) Proof. The proof is similar to the one for the orthogonal polynomials on the real line, that can be found, e.g., [13, equation (3.5)], and following discussion. ■ Corollary 2.2. The ratio of two consecutive Toeplitz determinants is expressed as Dℓ−1 Dℓ = κ2ℓ , ℓ ≥ 0. (2.4) Proof. Thanks to formula (2.3), we have that pℓ(z) = 1√ DℓDℓ−1 det  φ0 φ−1 . . . φ−ℓ+1 φ1 φ0 . . . φ−ℓ+2 ... ... . . . ... φℓ−1 φℓ−2 . . . φ0  zℓ + · · · = √ Dℓ−1 Dℓ zℓ + · · · , and by definition pℓ(z) = κℓπℓ(z) with the latter being the ℓ-th monic orthogonal polynomial on S1. Thus formula (2.4) follows. ■ Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 11 2.2 Riemann–Hilbert problem associated to OPUC The family {πn} of orthogonal polynomials has a well-known characterization in terms of a 2×2 dimensional Riemann–Hilbert problem, also depending on n ≥ 0. Riemann–Hilbert Problem 2.3. The function Y (z) := Y (n, θj ; z) : C → GL(2,C) has the following properties: (1) Y (z) is analytic for every z ∈ C \ S1; (2) Y (z) has continuous boundary values Y±(z) while approaching non-tangentially S1 either from the left or from the right, and they are related for all z ∈ S1 through Y+(z) = Y−(z)JY (z), with JY (z) = ( 1 z−new(z) 0 1 ) ; (3) Y (z) is normalized at ∞ as Y (z) ∼ ( I + ∞∑ j=1 Yj(n, θj) zj ) znσ3 , z → ∞, where σ3 denotes the Pauli’s matrix σ3 := ( 1 0 0 −1 ) . It is known from [3] that the above Riemann–Hilbert problem, for each n ≥ 0, admits a unique solution which is explicitly written in terms of the family {πn(z)}. Before stating the result, we introduce the following notation. For every polynomial q(z), z ∈ C, its reverse polynomial q∗(z) is defined as the polynomial of the same degree such that q∗(z) := znq ( z̄−1 ) . For every ( Lp ( S1 )) function f(y), its Cauchy transform Cf(z) is defined for any z /∈ S1 as (Cf(y)) (z) := 1 2πi ∫ S1 f(y) y − z dy. Remark 2.4. Notice that the results in [3] for the Riemann–Hilbert characterization a family of orthogonal polynomials on the unit circle are a sort of extension of the results known from [15, 20] for the case of orthogonal polynomials on the real line. Theorem 2.5. For every n ≥ 0, the Riemann–Hilbert Problem 2.3 admits a unique solution Y (z) that is written as Y (z) = ( πn(z) C ( y−nπn(y)e w(y) ) (z) −κ2n−1π ∗ n−1(z) −κ2n−1C ( y−nπ∗ n−1(y)e w(y) ) (z) ) . (2.5) Moreover, det(Y (z)) ≡ 1. Proof. See [3, Lemma 4.1]. ■ The solution Y (z) has a symmetry which will be very useful in the following section. Corollary 2.6. The unique solution Y (z) of the Riemann–Hilbert Problem 2.3 is such that Y (z) = σ3Y (0)−1Y ( z−1 ) znσ3σ3, (2.6) Y (z) = Y (z̄). (2.7) 12 T. Chouteau and S. Tarricone Proof. See [4, Proposition 5.12]. ■ Notice that the factor Y (0) = Y (n, θj ; 0) appearing in equation (2.6) has a very explicit form by equation (2.5). This will be useful in the following sections. Lemma 2.7. For every n ≥ 0, we have Y (0) = Y (n, θj ; 0) = ( xn κ−2 n −κ2n−1 xn ) , (2.8) where we denoted with xn := πn(0), and κn is defined as in equation (2.2). Moreover, we have κ2n−1 κ2n = 1− x2n, (2.9) and we have xn ∈ R. Proof. The first column of Y (n; 0) directly follows from the evaluation in z = 0 of Y (n; z) as given in equation (2.5). Indeed, Y 11(n; 0) = πn(0) and Y 21(n; 0) = −κ2n−1π ∗ n−1(0) but we observe that π∗ n−1(0) = zn−1πn−1 ( z̄−1 )∣∣ z=0 = zn−1 ( z−(n−1) + · · ·+ πn−1(0) )∣∣ z=0 = 1. Thus we conclude that Y 21(n; 0) = −κ2n−1. For what concerns the second column of Y (n; 0), we first find the (2, 2)-entry. This is indeed easily deduced from the symmetry given in (2.6). In the limit for z → ∞ it gives Y (n; 0) = σ3Y −1(n; 0)σ3, thus Y 22(n; 0) = Y 11(n; 0) = πn(0). Finally, for the entry (1, 2) of Y (n; 0), we compute it explicitly using the orthonormality property of the polynomials pm(z) Y 12(n; 0) = 1 2πi ∫ S1 πn(s)s −nw(s) s ds = ∫ π −π πn ( eiθ ) einθw ( eiθ )dθ 2π = 1 κ2n ∫ π −π pn ( eiθ ) pn ( eiθ ) w ( eiθ )dθ 2π = 1 κ2n . Equation (2.9) comes from the fact that det(Y (n, θj ; z)) = 1 identically in z and so in particular for z = 0 by writing Y (n, θj ; 0) as in equation (2.8), relation (2.9) is obtained. Finally, the fact that xn is real follows from the entry (1, 1) of equation (2.7) together with equation (2.5). ■ At this point, we are already able to express the ratio of Toeplitz determinants in terms of the constant term of the monic orthogonal polynomials, as follows. Corollary 2.8. For every n ≥ 1, the Toeplitz determinants Dn satisfy the recursion relation Dn−2Dn D2 n−1 = 1− x2n. (2.10) Proof. Putting together equation (2.9) with equation (2.4) (for two consecutive integers) we obtain the recursion relation (2.10). ■ Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 13 We emphasize again that the symbol φ(z) actually depends on the natural parameter N , so the Toeplitz determinants Dn, n ≥ 1 (1.3) do as well as xn = πn(0), n ≥ 1 do (since it is the constant coefficient of the n-th monic OPUC w.r.t. the N -depending measure (2.1), (1.1)). The N -dependence of the latter will be emphasized in the following section, where xn is proved to be a solution of the N -th higher order generalization of the discrete Painlevé II equation. We consider now the following matrix-valued function Ψ(n, θj ; z) := ( 1 0 0 κ−2 n ) Y (n, θj ; z) ( 1 0 0 zn ) ew(z) σ3 2 . (2.11) Thanks to the properties of Y (z;n, θj) from the Riemann–Hilbert Problem 2.3 one can prove that Ψ(n, θj ; z) satisfies the following Riemann–Hilbert problem. Riemann–Hilbert Problem 2.9. The function Ψ(z) := Ψ(n, θj ; z) : C → GL(2,C) has the following properties: (1) Ψ(z) is analytic for every z ∈ C \ { S1 ∪ {0} } ; (2) Ψ(z) has continuous boundary values Ψ±(z) while approaching non-tangentially S1 either from the left or from the right, and they are related for all z ∈ S1 through Ψ+(z) = Ψ−(z)J0, J0 = ( 1 1 0 1 ) ; (2.12) (3) Ψ(z) has asymptotic behavior near 0 given by Ψ(z) ∼ ( 1 0 0 κ−2 n ) Y (0) ( I + ∞∑ j=1 zj Ỹj(n) )( 1 0 0 zn ) ew(z) σ3 2 , z → 0; (2.13) (4) Ψ(z) has asymptotic behavior near ∞ given by Ψ(z) ∼ ( 1 0 0 κ−2 n )( I + ∞∑ j=1 Yj(n) zj )( zn 0 0 1 ) ew(z) σ3 2 , |z| → ∞. (2.14) Proposition 2.10. The function Ψ(n, θj ; z) defined in (2.11) solves the Riemann–Hilbert Prob- lem 2.9. Proof. The analyticity condition and the asymptotic expansions at 0, ∞ given in (2.13), (2.14) follows directly from the definition (2.11) and the fact that Y (z) solves the Riemann–Hilbert Problem 2.3. Condition (2.12) follows from direct computation Ψ(z)+ = ( 1 0 0 κ−2 n ) Y+(z) ( 1 0 0 zn ) ew(z) σ3 2 = ( 1 0 0 κ−2 n ) Y−(z)JY (z) ( 1 0 0 zn ) ew(z) σ3 2 = Ψ−(z) ( 1 0 0 z−n ) e−w(z) σ3 2 ( 1 z−new(z) 0 1 )( 1 0 0 zn ) ew(z) σ3 2 = Ψ−(z) ( 1 1 0 1 ) . ■ 2.3 A linear differential system for Ψ(z) From the solution of the Riemann–Hilbert Problem 2.9, we deduce the following equations (in the following we omit in Ψ the dependence on θj that should be considered only as parameters and not actual variables like n, z). 14 T. Chouteau and S. Tarricone Proposition 2.11. We have Ψ(n+ 1; z) = U(n; z)Ψ(n; z), ∂zΨ(n; z) = T (n; z)Ψ(n; z) (2.15) with U(n; z) := ( z + xnxn+1 −xn+1 − ( 1− x2n+1 ) xn 1− x2n+1 ) = σ+z + U0(n), (2.16) where σ+ := ( 1 0 0 0 ) and T (n; z) := T1(n)z N−1 + T2(n)z N−2 + · · ·+ T2N+1(n)z −N−1 = 2N+1∑ k=1 Tkz N−k, (2.17) where T1(n) = θN 2 σ3. (2.18) Remark 2.12. The coefficient (Ti(n))2⩽i⩽2N+1 defined in equation (2.17) will be computed in Section 3. Proof. We first prove the first equation. We start by defining the quantity U(n; z) := Ψ(n+ 1; z)Ψ−1(n; z). Since the jump condition for Ψ(z) (2.12) is independent of n, U(n; z) is analytic everywhere. Plugging in equation (2.14), we have the expansion at ∞ U(n; z) = ( 1 0 0 κ−2 n+1 )( I + Y1(n+ 1) z +O ( z−2 )) z(n+1)σ3 ( 1 0 0 z ) z−nσ3 × ( I − Y1(n) z +O ( z−2 ))(1 0 0 κ2n ) , from which we deduce that U(n; z) is a polynomial in z of degree 1, by Liouville theorem. Moreover, its matrix-valued coefficient are written as U(n; z) = z ( 1 0 0 0 ) + ( 1 0 0 κ−2 n+1 ) Y (n+ 1; 0) ( 1 0 0 0 ) Y −1(n; 0) ( 1 0 0 κ2n ) =U0(n) . Doing the computation and using equation (2.8), we obtain U0(n) = ( Y 11(n+ 1; 0)Y 22(n; 0) −κ2nY 11(n+ 1; 0)Y 12(n; 0) κ−2 n+1Y 21(n+ 1; 0)Y 22(n, 0) −Y 21(n+ 1; 0)Y 12(n; 0) ) = ( xn+1xn −xn+1 − ( 1− x2n+1 ) xn 1− x2n+1 ) . For what concerns the second equation, we define T (n; z) := ∂zΨ(n; z)Ψ−1(n; z). From the asymptotic behavior of Ψ(n; z) at 0 and ∞, we can deduce that T (n; z) is a meromorphic function in z with behavior at ∞ described by T (n; z) ∼ ( 1 0 0 κ−2 n )( I + Y1(n) z +O ( z−2 ))V ′(z) 2 σ3 ( I − Y1(n) z +O ( z−2 ))(1 0 0 κ2n ) Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 15 (polynomial behavior of degree N − 1) while at 0 its behavior is described by T (n; z) ∼ ( 1 0 0 κ−2 n ) Y (n, 0) ( I + Ỹ1(n)z +O ( z2 )) × −V ′(z−1) 2z2 σ3 ( I − Ỹ1(n)z +O ( z2 ))(1 0 0 κ2n ) , i.e., there is a pole of order N + 1. In conclusion, we can write T (n; z) = θN 2 σ3z N−1 + T2(n)z N−2 + · · ·+ T2N+1(n)z −N−1. ■ Moreover, thanks to the symmetry for the solution of the Riemann–Hilbert problem Y (z) stated in (2.6), we have that the coefficient matrix T (n; z) satisfies a symmetry property. Proposition 2.13. T (n; z) has the following symmetry: T (n; z−1) = −z2 ( K(n)T (n; z)K(n)−1 − nz−1I2 ) (2.19) with K(n) := ( 1 0 0 κ−2 n ) Y (n; 0)σ3 ( 1 0 0 κ2n ) . Remark 2.14. Notice that for all n, the matrix K(n) is s.t. K(n)−1 = K(n) since we have the identity x2n + κ2 n−1 κ2 n = 1. Proof. On the one hand, ∂z ( Ψ ( n; z−1 )) = − 1 z2 T ( n; z−1 ) Ψ ( n; z−1 ) . On the other hand, using the symmetry (2.6) for Y we deduce the following symmetry for Ψ: Ψ ( n; z−1 ) = z−n ( 1 0 0 κ−2 n ) Y (0)σ3 ( 1 0 0 κ2n ) Ψ(n; z)σ3. This previous equation leads to ∂z ( Ψ ( n; z−1 )) = z−n ( 1 0 0 κ−2 n ) Y (0)σ3 ( 1 0 0 κ2n ) ∂zΨ(n; z)σ3 − nz−1Ψ ( n; z−1 ) . Then T ( n; z−1 ) = − z2 (( 1 0 0 κ−2 n ) Y (0)σ3 ( 1 0 0 κ2n ) T (n; z) ( 1 0 0 κ−2 n ) σ3Y (0)−1 × ( 1 0 0 κ2n ) − nz−1I2 ) . ■ The symmetry (2.19) reflects on the coefficients Tk(n), k = 1, . . . , 2N + 1 as written below. Corollary 2.15. The coefficients Tk(n), k = 1, . . . , 2N + 1 satisfy Tj(n) = −K(n)T2N+2−j(n)K(n)−1, j = 1, . . . , N, (2.20) TN+1(n) = −K(n)TN+1(n)K(n)−1 + nI2. (2.21) 16 T. Chouteau and S. Tarricone Proof. Indeed, by replacing the exact shape of T (n; z) in equation (2.19), we have 2N+1∑ k=1 Tk(n)z −N+k = T ( n; z−1 ) = −z2 ( 2N+1∑ k=1 KTk(n)K −1zN−k − nz−1I2 ) = − 2N+1∑ k=1 KTk(n)K −1zN+2−k + nzI2 = − 2N+1∑ j=1 KT2N+2−j(n)K −1z−N+j + nzI2, so looking at the powers z−N+j for j = 1, . . . , N , we get equation (2.20) and for j = N + 1, we get equation (2.21). ■ Notice first that from equations (2.20) if the first N + 1 coefficients of T (n; z) are known, then we can obtain the remaining ones. Second, notice that the coefficient TN+1(n) plays an important role since it solves an equation, the one given in (2.21). 2.4 Relation with the Cresswell–Joshi Lax pair To conclude this section, we describe how the Lax pair (2.15) is related with the one of the discrete Painlevé II hierarchy (1.16) originally introduced by Cresswell and Joshi in [11] as follows. Definition 2.16. A Lax pair for the discrete Painlevé II hierarchy is given by a pair of matrices (Ln(z),Mn(z)), defining the coefficients of a discrete-differential system for a matrix-valued function Φ(n; z), such as Φ(n+ 1; z) = ( z xn xn 1/z ) Φ(n; z) = Ln(z)Φ(n; z), (2.22) ∂ ∂z Φ(n; z) = Mn(z)Φ(n; z), (2.23) with the property that Mn(z) = ( An(z) Bn(z) Cn(z) −An(z) ) with An, Bn and Cn are rational in z (and depending also on N). Remark 2.17. Specifically, in [11, Section 3.1], the authors proved that the compatibility condition of the system of equations (2.22) and (2.23) defines the coefficients of the matrixMn(z), leaving in turns only one discrete equation of order 2N for xn. This is defined as the N -th member of the discrete Painlevé II hierarchy. We establish now a link between this Lax Pair and the system (2.15) we obtained starting from the OPUC. We define Φ(n; z) := σ3 ( z−n+3/2 0 0 z−n+1/2 )( 1 0 −xn−1 1 ) Ψ ( n− 1; z2 ) . Proposition 2.18. Φ(n; z) defined as above satisfies the system of equations (2.22) and (2.23). Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 17 Proof. First we compute the discrete equation for Φ(n; z). From the definition, we have Φ(n+ 1; z) = σ3 ( z−n+1/2 0 0 z−n−1/2 )( 1 0 −xn 1 ) Ψ ( n; z2 ) . According to equation (2.15), Φ(n+ 1; z) = σ3 ( z−n+1/2 0 0 z−n−1/2 )( 1 0 −xn 1 ) U ( n− 1; z2 ) Ψ ( n− 1; z2 ) = σ3 ( z−n+1/2 0 0 z−n−1/2 )( 1 0 −xn 1 ) U ( n− 1; z2 )( 1 0 xn−1 1 ) × ( zn−3/2 0 0 zn−1/2 ) σ3Φ(n; z) = ( z xn xn 1/z ) Φ(n; z). Now we compute the derivative with respect to z. Defining Mn(z) := ( ∂ ∂zΦ(n; z) ) Φ(n; z)−1, similar computations lead to Mn(z) = z−1σ3 ( −n+ 3/2 0 0 −n+ 1/2 ) σ3 + 2zσ3 ( z 0 0 1 )( 1 0 −xn−1 1 ) × T ( n− 1; z2 )( 1 0 xn−1 1 )( z−1 0 0 1 ) σ3. (2.24) We need to prove two things: first the trace of Mn(z) is null and then entries of Mn(z) are rational in z. For the trace of Mn(z) we use the fact that Tr(T (n; z)) = nz−1. Then Tr(Mn(z)) = (−2n+ 2)z−1 + 2zTr ( T ( n− 1; z2 )) = 0. From the expression of T (n; z) (2.17) and the equation (2.24), we conclude entries of Mn(z) are rational in z. ■ 3 From the Lax Pair to the discrete Painlevé II hierarchy In this section, we study the compatibility condition associated to the linear system (2.15). This first allows us to reconstruct completely the matrix T (n; z) and then to obtain an explicit 2N order discrete equation for xn which corresponds to equation (1.10). 3.1 The symmetry in the compatibility condition We study the consequences of the symmetry (2.19) for the matrix T (n; z) on the compatibil- ity condition for the Lax pair introduced in Proposition 2.11. More precisely, we show that, thanks to the symmetry (2.19), the compatibility condition contains an overdetermined system of equations. We recall that the compatibility condition reads as σ+ − T (n+ 1; z)U(n; z) + U(n; z)T (n; z) = 0, (3.1) where we have to replace U(n; z) as in (2.16) and T (n; z) as T (n; z) = N+1∑ k=1 Tk(n)z N−k + 2N+1∑ k=N+2 −K(n)T2N+2−k(n)K(n)−1zN−k, (3.2) and with the coefficient TN+1(n) satisfying equation (2.21). 18 T. Chouteau and S. Tarricone Lemma 3.1. The compatibility condition (3.1), for U(n; z), T (n; z) as described above, corre- sponds to the following system T1(n+ 1)σ+ − σ+T1(n) = 0, Tj+1(n+ 1)σ+ − σ+Tj+1(n) + Tj(n+ 1)U0(n)− U0(n)Tj(n) = σ+δj,N , j = 1, . . . , N, TN+1(n) = −K(n)TN+1(n)K(n)−1 + nI2. Proof. The compatibility condition (3.1), after replacing U(n; z), T (n; z) of the prescribed form, involves powers of z from N to −N − 1. Imposing that the coefficients of each of these powers of z is identically zero, we obtain the following equations: zN : T1(n+ 1)σ+ − σ+T1(n) = 0, (3.3) zN−j , j = 1, . . . , N : Tj+1(n+ 1)σ+ − σ+Tj+1(n) + Tj(n+ 1)U0(n)− U0(n)Tj(n) = σ+δj,N, (3.4) z−1 : TN+1(n+ 1)U0(n)− U0(n)TN+1(n)−K(n+ 1)TN (n+ 1)K(n+ 1)−1σ+ + σ+K(n)TN (n)K(n)−1 = 0, (3.5) zN−j , j = N + 2, . . . , 2N : −K(n+ 1)T2N+1−j(n+ 1)K(n+ 1)−1σ+ + σ+K(n)T2N+1−j(n)K(n)−1 + U0(n)K(n)T2N+2−j(n)K(n)−1 −K(n+ 1)T2N+2−j(n+ 1)K(n+ 1)−1U0(n) = 0, (3.6) z−N−1 : −K(n+ 1)T1(n+ 1)K(n+ 1)−1U0(n) + U0(n)K(n)T1(n)K(n)−1 = 0. (3.7) With the change of indices 2N+1−j = k ⇐⇒ k = 2N+1−j = N−1, . . . , 1, the equation (3.6) becomes: −K(n+ 1)Tk(n+ 1)K(n+ 1)−1σ+ + σ+K(n)Tk(n)K(n)−1 −K(n+ 1)Tk+1(n+ 1)K(n+ 1)−1U0(n) + U0(n)K(n)Tk+1(n)K(n)−1 = 0. (3.8) We now show that equations (3.5), (3.6), (3.7) are equivalent to the first ones (3.3), (3.4) thanks to the symmetry of the coefficients Tk(n) given in (2.20) together with the equation for TN+1(n), already obtained in (2.21). To start with, we notice the following relations: Ũ0(n) := K(n+ 1)−1U0(n)K(n) = σ+, σ̃(n) := K(n+ 1)−1σ+K(n) = U0(n), deduced by using multiple times relation (2.9), namely x2n + κ2 n−1 κ2 n = 1. 1. Let us consider first the equation (3.7) obtained from the coefficient of the term z−N−1. Multiplying by K(n+ 1)−1 to the left and by K(n) to the right, we obtain −T1(n+ 1)Ũ0(n) + Ũ0(n)T1(n) = 0, that is exactly (3.3). 2. Let us consider now equations (3.8), obtained from the coefficients of the term zN−j , j = N + 2, . . . , 2N . By multiplying by K(n+ 1)−1 to the left and by K(n) to the right as before, we obtain the equations for k = N − 1, . . . , 1 −Tk(n+ 1)σ̃(n) + σ̃(n)Tk(n)− Tk+1(n+ 1)Ũ0(n) + Ũ0(n)Tk+1(n) = 0, which is exactly equation (3.4) for j = 1, . . . , N − 1. Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 19 3. The last equation is (3.5) obtained from the coefficient of the term z−1. We multiply, again, by K(n+ 1)−1 to the left and by K(n) to the right, and we get K(n+ 1)−1TN+1(n+ 1)K(n+ 1)Ũ0(n)− Ũ0(n)K(n)−1TN+1(n)K(n) − TN (n+ 1)σ̃(n) + σ̃(n)TN (n) = 0, and then we replace the symmetry for the term TN+1(n) namely the equation (2.21) (that indeed it has not be used until now) −TN+1(n+ 1)Ũ0(n) + Ũ0(n)TN+1(n) + Ũ0(n)− TN (n+ 1)σ̃(n) + σ̃(n)TN (n) = 0. And this is again exactly equation (3.4), for j = N . Thus the compatibility condition (3.1) is reduced to the equations in the statement, namely equations (3.3), (3.4), (2.21). ■ Now, we use equations (3.3), (3.4) together with the initial condition for T1(n) given in (2.18), to recursively find the coefficients Tk(n), for k = 1, . . . , N+1, in terms of the xn±j , j = 1, . . . , N . With the coefficients Tk(n) computed in such a way, the symmetry for TN+1(n), i.e., equa- tion (2.21), once TN+1(n) is determined, provides an actual discrete equation for xn of order 2N , that is what we call the higher order analogue of the discrete Painlevé II equation (that coincide for N = 1, 2 to the ones already appeared in [1, 6, 11]). 3.2 The recursion In this subsection, we explain how equations (3.3), (3.4) resulting from the compatibility con- dition (3.1) can be used to find recursively (in k) all the coefficients Tk(n), k = 1, . . . , N + 1 of T (n; z). Lemma 3.2. For every i = 1, . . . , N , starting from the initial condition (2.18) T1(n) = θN 2 σ3, we have Ti+1,12(n) = xn+1 ( 2∆−1 + I )(xn+1 vn+1 Ti,21(n+ 1)− xnTi,12(n) ) + vn+1Ti,12(n+ 1) − xnxn+1Ti,12(n), Ti+1,21(n+ 1) = xnvn+1 ( 2∆−1 + I )(xn+1 vn+1 Ti,21(n+ 1)− xnTi,12(n) ) + vn+1Ti,21(n) − xnxn+1Ti,21(n+ 1), Ti+1,11(n) = −Ti+1,22(n) + nδi,N = ∆−1 ( −xn+1 vn+1 Ti+1,21(n+ 1) + xnTi+1,12(n) ) + nδi,N , where ∆: Ti(n) → Ti(n+ 1)− Ti(n), (3.9) vn := 1− x2n, (3.10) Proof. We rewrite equations (3.3), (3.4) for i = 1, . . . , N , entry by entry. For the first one, we have{ T1,11(n+ 1)− T1,11(n) = 0, T1,12(n) = T1,21(n+ 1) = 0. 20 T. Chouteau and S. Tarricone This is satisfied by T1(n) given in (2.18). For the second one, for any 1 ⩽ i ⩽ N we have the four equations: Ti+1,11(n+ 1)− Ti+1,11(n) = −Ti,11(n+ 1)xnxn+1 + Ti,12(n+ 1) ( 1− x2n+1 ) xn + xnxn+1Ti,11(n)− xn+1Ti,21(n) + δi,N , Ti+1,12(n) = −xn+1Ti,11(n+ 1) + Ti,12(n+ 1) ( 1− x2n+1 ) − xnxn+1Ti,12(n) + xn+1Ti,22(n), Ti+1,21(n+ 1) = −Ti,21(n+ 1)xnxn+1 + Ti,22(n+ 1)xn ( 1− x2n+1 ) − Ti,11(n)xn ( 1− x2n+1 ) + ( 1− x2n+1 ) Ti,21(n), 0 = Ti,21(n+ 1)xn+1− Ti,22(n+ 1) ( 1− x2n+1 ) − xn ( 1− x2n+1 ) Ti,12(n) + Ti,22(n) ( 1− x2n+1 ) . Using the notations introduced in (3.9), (3.10), the previous equations with 1 ⩽ i ⩽ N become ∆Ti+1,11(n) = −xnxn+1∆Ti,11(n) + xnvn+1Ti,12(n+ 1)− xn+1Ti,21(n) + δi,N , (3.11) Ti+1,12(n) = −xn+1Ti,11(n+ 1)+ vn+1Ti,12(n+ 1)− xnxn+1Ti,12(n)+ xn+1Ti,22(n), (3.12) Ti+1,21(n+ 1) = −xnxn+1Ti,21(n+ 1) + xnvn+1Ti,22(n+ 1)− xnvn+1Ti,11(n) + vn+1Ti,21(n), (3.13) vn+1∆Ti,22(n) = xn+1Ti,21(n+ 1)− xnvn+1Ti,12(n). (3.14) From these equations, we see that in order to obtain the diagonal terms, there is a “discrete integration” to perform, while the off-diagonal terms are directly determined from the previous ones. Moreover, we can rewrite the four equation as only two equations involving only the off- diagonal terms. Indeed, because of Tr(T (n; z)) = nz−1, Ti,11(n, z) = −Ti,22(n, z) for 1 ⩽ i ⩽ N . Thus (3.14) can be written as vn+1∆Ti,11(n) = −xn+1Ti,21(n+ 1) + xnvn+1Ti,12(n). Formally, 1 ⩽ i ⩽ N , Ti,11(n) = −Ti,22(n) = ∆−1 ( −xn+1 vn+1 Ti,21(n+ 1) + xnTi,12(n) ) , (3.15) which still holds for i = N +1 up to adding the “constant” n on the right hand side. Using this in (3.12) and (3.13), we obtain: Ti+1,12(n) = xn+1 ( 2∆−1 + I )(xn+1 vn+1 Ti,21(n+ 1)− xnTi,12(n) ) + vn+1Ti,12(n+ 1) − xnxn+1Ti,12(n), Ti+1,21(n+ 1) = xnvn+1 ( 2∆−1 + I )(xn+1 vn+1 Ti,21(n+ 1)− xnTi,12(n) ) + vn+1Ti,21(n) − xnxn+1Ti,21(n+ 1). ■ We notice that, defining the discrete recursion operator L ( un yn ) =  xn+1 ( 2∆−1 + I )(xn+1 vn+1 yn − xnun ) + (vn+1(∆ + I)− xnxn+1)un xnvn+1 ( 2∆−1+I )(xn+1 vn+1 yn−xnun ) + ( vn+1(∆ + I)−1−xnxn+1 ) yn , (3.16) Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 21 we can rewrite the two equations for the off-diagonal entries of Ti(n) obtained above as( Ti+1,12(n) Ti+1,21(n+ 1) ) = L ( Ti,12(n) Ti,21(n+ 1) ) , 1 ⩽ i ⩽ N. (3.17) And, recursively we obtain( TN+1,12(n) TN+1,21(n+ 1) ) = LN ( 0 0 ) . (3.18) This procedure allows to construct the whole matrix T (n; z), starting from the initial condition T1(n) = θN 2 σ3 and iterating the operator L we obtain off diagonal terms of T (n; z) and compute diagonal one with equation (3.15). Below we implemented this method to find the matrix T (n; z) in the first few cases N = 1, 2. Example 3.3. In the case N = 1, the matrix T (n; z) = T1(n)+T2(n)z −1+T3(n)z −2. Knowing T1(n), we only have to find T2(n) using the recurrence relation given from the compatibility, i.e., equations (3.11), (3.12), (3.13) for i = 1. Since: T1,12(n) = T1,21(n) = 0, and T1,11(n) = θN/2 = −T1,22(n), we have T2,11(n) = n, T2,12(n) = −xn+1(T1,11(n+ 1) + T1,11(n)) = −θ1xn+1, T2,21(n+ 1) = xnvn+1(T1,22(n+ 1) + T1,22(n)) = −θ1xnvn+1, and T2,22(n) = n−T2,11(n) = 0. Moreover, the symmetry which reflects terms of T (n; z) two by two gives T3(n) = −K(n)T1(n)K(n). Thus the Lax matrix for N = 1 is T (n; z) = θ1 2 ( 1 0 0 −1 ) + 1 z ( n −θ1xn+1 −θ1vnxn−1 0 ) + θ1 z2 ( 1 2 − x2n xn vnxn x2n − 1 2 ) . Example 3.4. In the case N = 2, the matrix T (n; z) = T1(n)z+T2(n)+T3(n)z −1+T4(n)z −2+ T5(n)z −3. This time we have to find T2(n) (that will be almost the same as before) and also T3(n) using the recurrence relation given from the compatibility, i.e., equations (3.11), (3.12), (3.13) for i = 1 and 2. First we find T2(n) (i = 1 above), we have T2,11(n) = θ1 2 , T2,12(n) = −xn+1(T1,11(n+ 1) + T1,11(n)) = −θ2xn+1, T2,21(n+ 1) = xnvn+1(T1,22(n+ 1) + T1,22(n)) = −θ2xnvn+1, and T2,22(n) = −T2,11 = − θ1 2 . Then we consider the equation for i = 2 and find T3(n). We have ∆T3,11(n) = xnvn+1(−θ2xn+2)− xn+1(−θ2xn−1vn) + 1 =⇒ T3,11(n) = n− θ2xn−1xn+1vn, T3,12(n) = −θ1xn+1 − θ2 ( vn+1xn+2 − xnx 2 n+1 ) , T3,21(n+ 1) = ( −θ1xn − θ2 ( vnxn−1 − x2nxn+1 )) vn+1, T3,22(n) = n− T3,11(n) = θ2xn−1xn+1vn. Finally, we take T4(n) = −K(n)T2(n)K(n) and T5(n) = −K(n)T1(n)K(n). Thus the Lax matrix for N = 2 is T (n; z) = z θ2 2 ( 1 0 0 −1 ) + ( θ1 2 −θ2xn+1 −θ2xn−1vn − θ1 2 ) 22 T. Chouteau and S. Tarricone + 1 z ( n− θ2xn−1xn+1vn −θ1xn+1 − θ2 ( vn+1xn+2 − xnx 2 n+1 )( −θ1xn−1 − θ2 ( vn−1xn−2 − xnx 2 n−1 )) vn θ2xn−1xn+1vn ) + 1 z2 ( −θ2vn(xnxn−1 + xnxn+1) + θ1 2 ( vn − x2n ) −θ2 ( vnxn−1 + x2nxn+1 ) −θ2 ( vnxn+1 + x2nxn−1 ) vn θ2vn(xnxn−1 + xnxn+1)− θ1 2 ( vn − x2n )) + θ2 z3 ( 1 2 − x2n xn vnxn x2n − 1 2 ) . Now that we have reconstructed the whole matrix T (n; z) in terms of xn±j , j = −N, . . . , N we are left with the equation that TN+1(n) has to satisfy, namely (2.21). We now show that actually this coincide with only one scalar equation in TN+1,12 and TN+1,21. Indeed, entry by entry it reads as the following system of four equations. From the off-diagonal entries vnTN+1,12(n) = xn(TN+1,11(n)− TN+1,22(n))− TN+1,21(n), vnTN+1,21(n) = xnvn(TN+1,11(n)− TN+1,22(n))− v2nTN+1,12(n) (3.19) and from the diagonal entries n− ( 1 + x2n ) TN+1,11(n)− vnTN+1,22(n) + xnTN+1,21(n) + xnvnTN+1,12(n) = 0, n− ( 1 + x2n ) TN+1,22(n)− vnTN+1,11(n)− xnTN+1,21(n)− xnvnTN+1,12(n) = 0. We notice first that the four above equations are all the same. The first and the second equations are the same up to a multiplication by vn. Using the relation TN+1,11(n) + TN+1,22(n) = n, we can rewrite the third and the forth equations and obtain the same equation up to a sign. Finally, multiplying by xn the first equation and using the relation TN+1,11(n)+TN+1,22(n) = n we obtain the third one. Thus from now on we will refer only to (3.19), as for the remaining equation. Using equation (3.14) and Tr(T (n; z)) = nz−1, we express equation (3.19) in function of TN+1,12(n) and TN+1,21(n). Consider equation (3.19), with the identity Tr(TN+1(n)) = n, it is rewritten as vnTN+1,12(n) = xn(n− 2TN+1,22(n))− TN+1,21(n). Equation (3.14) holds also for i = N + 1. It means it is possible to replace TN+1,22(n) in the previous equation and obtain nxn − vnTN+1,12(n)− TN+1,21(n) − 2xn∆ −1 ( −xnTN+1,12(n) + xn+1 vn+1 (∆ + I)TN+1,21(n) ) = 0. (3.20) 3.3 The relation between Ti,12(n) and Ti,21(n) The previous equation (3.2) depends on TN+1,12(n) and TN+1,21(n). The aim of this part is to establish a connection between Ti,12(n) and Ti,21(n) to rewrite equation (3.2) just in function of TN+1,12(n). To accomplish this, we study the compatibility condition of C(n; z) := T (n; z)2 and U(n; z). C(n; z) is rational in z with a pole of order −2N − 2 at 0. We write C(n; z) as C(n; z) = 4N+1∑ i=1 Ci(n)z 2N−1−i (3.21) with Ci(n) := i∑ j=1 Tj(n)Ti+1−j(n) (3.22) where C1(n) = θ2N 4 I2. Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 23 In what follows we will need the following lemma: Lemma 3.5. Diagonal coefficients of Ci(n) defined as in (3.22) satisfy the following equation: ∀1 ⩽ i ⩽ N, Ci,11(n) = Ci,22(n), CN+1,11(n) = nθN + CN+1,22(n). Proof. We express Ci,11(n) in function of Ti,kj(n). With the equation (3.22) Ci,11(n) = i∑ j=1 Tj,11(n)Ti+1−j,11(n) + Tj,12(n)Ti+1−j,21(n). Then, the sum index change j = i− k + 1 leads to Ci,11(n) = i∑ k=1 Ti−k+1,11(n)Tk,11(n) + Ti−k+1,12(n)Tk,21(n). Finally, with the relation Tr(T (n; z)) = nz−1, � if 1 ⩽ i ⩽ N , Ci,11(n) = i∑ k=1 Ti−k+1,22(n)Tk,22(n) + Tk,21(n)Ti−k+1,12(n) = Ci,22(n). � if i = N + 1, CN+1,11(n) = −2nT1,22(n) + N+1∑ k=1 TN−k+2,22(n)Tk,22(n) + Tk,21(n)TN−k+2,12(n) = nθN + CN+1,22(n). ■ We deduce the compatibility condition for C and U from the one for T and U . Lemma 3.6. C(n; z) (3.21) and U(n; z) (2.16) satisfy the following compatibility condition: C(n+ 1; z)U(n; z)− U(n; z)C(n; z) = T (n+ 1; z)σ+ + σ+T (n; z). (3.23) Proof. Multiplying on the left (resp. on the right) equation (3.1) by T (n+1; z) (resp. T (n; z)) and summing these two equations leads to the result. ■ The left (resp. right) hand side of the equation in the previous lemma is an expression in powers of z from z2N−1 to z−2N−2 (resp. from zN−1 to z−N−1). This equation leads to recursive equation for Ci(n). We consider only expression in powers of z from z2N−1 to zN−1. According to (3.1) and (3.23), ∀1 ⩽ i ⩽ N , Ci(n) and Ti(n) satisfy the same recursive equation (see equations (3.11)–(3.14)). For i = N + 1, the equation is a bit different. The term with δi,N is now multiplied by θN . From these equations we deduce the following result. Proposition 3.7. Let Ci(n) be as in (3.22). Then ∀1 ⩽ i ⩽ N , Ci(n) = αiI2 and CN+1(n) = θNnσ+ + αN+1I2. 24 T. Chouteau and S. Tarricone Proof. We prove Proposition 3.7 by induction. For i = 1, we already know C1(n) = θ2N 4 . Suppose Ci(n) = αiI2 for i ⩽ N − 1. Ci+1(n) satisfies the following equations: ∆Ci+1,11(n) = −xnxn+1∆Ci,11(n) + xnvn+1Ci,12(n+ 1)− xn+1Ci,21(n) + θNδi,N , Ci+1,12(n) = −xn+1Ci,11(n+ 1) + vn+1Ci,12(n+ 1)− xnxn+1Ci,12(n) + xn+1Ci,22(n), Ci+1,21(n+ 1) = −xnxn+1Ci,21(n+ 1) + xnvn+1Ci,22(n+ 1)− xnvn+1Ci,11(n) + vn+1Ci,21(n). Using induction hypothesis, ∆Ci+1,11(n) = −0 · xnxn+1 + 0 · xnvn+1 − 0 · xn+1 + θNδi,N = θNδi,N , Ci+1,12(n) = −xn+1αi + 0 · vn+1 − 0 · xnxn+1 + xn+1αi = 0, Ci+1,21(n+ 1) = −0 · xnxn+1 + xnvn+1αi − xnvn+1αi + 0 · vn+1 = 0. From the first equation, we conclude Ci+1,11(n) = αi+1 if i ⩽ N − 1 (resp. CN+1,11(n) = θNn+αN+1 if i = N) and according to Lemma 3.5 Ci+1,22(n) = αi+1 (resp. CN+1,22(n) = αN+1) which concludes the proof. ■ From equation (3.22) and Proposition 3.7, we obtain θNTi,11(n) = αi − i−1∑ j=2 Tj,11(n)Ti−j+1,11(n) + Tj,12(n)Ti−j+1,21(n), (3.24) θNTN+1,11(n) = nθN + αN+1 − N∑ j=2 Tj,11(n)TN−j+2,11(n) + Tj,12(n)TN−j+2,21(n). (3.25) With all this discussion on C(n; z) it is now possible to prove the following proposition. Proposition 3.8. The following holds: ∀1 ⩽ i ⩽ N + 1, Ti,11(n), Ti,12(n) and Ti,21(n) are polynomials in xn+j’s. Moreover, the following symmetries hold: ∃(Qi,n((un+j)1−i⩽j⩽i−1), Pi,n((un+j)1−i⩽j⩽i−1)) polynomials in un+j’s such that, Ti,11(n) = Qi,n((xn+j)1−i⩽j⩽i−1) = Qi,n((xn−j)1−i⩽j⩽i−1), Ti,12(n) = Pi,n((xn+j)1−i⩽j⩽i−1), Ti,21(n) = vnPi,n((xn−j)1−i⩽j⩽i−1). Proof. We prove this proposition by strong induction. For i = 1, T1(n) = θN 2 σ3, then defin- ing Q1,n(un) := θN 2 , P1,n(un) := 0; T1,11(n) = Q1,n(xn), T1,12(n) = P1,n(xn) and T1,21(n) = vnP1,n(xn). Now suppose the property true for all j ∈ [[1, i]] with i ⩽ N and let (Qj,n, Pj,n)j⩽i be polynomials in xn+j ’s satisfying the property. According to (3.24) (and (3.25) for i = N) and strong induction hypothesis, Ti+1(n) is a polynomial in xn+j ’s and the invariance when you exchange xn+j by xn−j holds. Because of equation (3.12) (resp. equation (3.13)) and of induction hypothesis, there exists Pi+1,n((un+j)−i⩽j⩽i) (resp. P̃i+1,n((un+j)−i⩽j⩽i)) a polynomial such that Ti+1,12(n) = Pi+1,n((xn+j)−i⩽j⩽i), Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 25 respectively, Ti+1,21(n) = P̃i+1,n((xn+j)−i⩽j⩽i). Now we establish the link between Pi+1,n and P̃i+1,n. According to equation (3.12) and the relation Tr(T (n; z)) = nz−1, Pi+1,n ( (xn+j) i j=−i ) = − xn+1Qi,n+1 ( (xn+j) i−2 j=−i ) + vn+1Pi,n+1 ( (xn+j) i−2 j=−i ) − xnxn+1Pi,n ( (xn+j) i−1 j=1−i ) − xn+1Qi,n ( (xn+j) i−1 j=1−i ) . Then vnPi+1,n ( (xn−j) i j=−i ) = vn ( − xn−1Qi,n−1 ( (xn−j) i−2 j=−i ) + vn−1Pi,n−1 ( (xn−j) i−2 j=−i ) − xnxn−1Pi,n ( (xn−j) i−1 j=1−i ) − xn−1Qi,n ( (xn−j) i−1 j=1−i )) . From induction hypothesis and Tr(T (n; z)) = nz−1 vnPi+1,n ( (xn−j) i j=−i ) = − xn−1vnTi,11(n− 1) + vnTi,21(n− 1) + xn−1xnTi,21(n) + xn−1vnTi,22(n). According to equation (3.13), vnPi+1,n ( (xn−j) i j=−i ) = Ti+1,21(n+ 1). Then vnPi+1,n ( (xn−j) i j=−i ) = P̃i+1,n ( (xn+j)−i⩽j⩽i ) and this concludes the proof. ■ Define C[(xj)j∈[[0,2n]]] and the transformation Permn : C[(xj)j∈[[0,2n]]] −→ C[(xj)j∈[[0,2n]]], P ((xn+j)−n⩽j⩽n) 7−→ P ((xn−j)−n⩽j⩽n). From the previous proposition, Ti,21(n) = vnPermn(Ti,12(n)). (3.26) Remark 3.9. As a consequence of the Proposition 3.8, the equation (3.19) is a polynomial in xn+j ’s and is invariant when you apply Permn to this equation because Perm2 n = Id and Permnvn = vnPermn. We use the link we established in Proposition 3.8 between Ti,12(n) and Ti,21(n) to rewrite the operator L (3.16) as a scalar operator: L(un) := ( xn+1 ( 2∆−1 + I ) ((∆ + I)xnPermn − xn) + vn+1(∆ + I)− xnxn+1 ) un. (3.27) Finally, collecting all the results from the previous sections, we state and proof the following theorem. Theorem 3.10. The system (2.15), with T (n; z) of the form (3.2) and coefficient TN+1(n) satis- fying the symmetry condition (2.21), is a Lax pair for the N -th higher order discrete Painlevé II equation and the equation is given by the expression: nxn + ( 2xn∆ −1(xn − (∆ + I)xnPermn)− vn − vnPermn ) TN+1,12(n) = 0, (3.28) where TN+1,12(n) = LN (0) with L as in (3.27). 26 T. Chouteau and S. Tarricone Proof. Replacing TN+1,21(n) with the relation (3.26), equation (3.2) now reads as nxn + ( 2xn∆ −1(xn − (∆ + I)xnPermn)− vn − vnPermn ) TN+1,12(n) = 0. Equations (3.17) and (3.18) with the relation (3.26) reduce to Ti+1,12(n) = L(Ti,12(n)) and TN+1,12(n) = LN (0), which concludes the proof. ■ The next two examples explain for N = 1, 2 how to compute explicitly equation (3.28). Example 3.11. Using the expression defined in Theorem 3.10, we compute the first equa- tion (1.13) and the second (1.14). For N = 1: First we compute T2,12(n) with the operator L (3.27): T2,12(n) = 2xn+1∆ −1(0) = −θ1xn+1, where −θ1/2 is the integration constant. Replacing T2,12(n) in equation (3.28), nxn + vnθ1(xn+1 + xn−1) + 2xn∆ −1(θ1xnxn+1 − θ1xnxn+1) = 0. Then (n+ α)xn + θ1vn(xn+1 + xn−1) = 0. This equation is the same as equation (1.13) if we choose the integration constant α to be zero. For N = 2: We compute T3,12(n). Computations are the same for T2,12(n) except for the integration constant, T2,12(n) = −θ2xn+1. T3,12(n) = L(T2,12(n)) = ( xnx 2 n+1 − vn+1xn+2 ) θ2 + xn+1 ( 2∆−1 + I ) (−θ2xnxn+1 + θ2xnxn+1) Then T3,12(n) = θ2 ( xnx 2 n+1 − vn+1xn+2 ) − θ1xn+1. Replacing T3,12(n) in equation (3.28), (n+ α)xn + θ2vn ( vn+1xn+2 + vn−1xn−2 − xn(xn+1 + xn−1) 2 ) + θ1vn(xn+1 + xn−1) = 0 which is the same equation as (1.14). We finally conclude the work by noticing that Theorem 3.10 together with Corollary 2.8 give the proof of Theorem 1.2. Remark 3.12. In our setting, the fixed N ≥ 1 define the order (2N) of the discrete equation solved by xn, the quantity related to the Toeplitz determinants Dn. An alternative approach could be to leave N variate and consider it as a second discrete variable for xn. In effect, this is done in [19], where the authors consider orthogonal polynomials on the real line, w.r.t. a weight ρ(λ;N)dλ and where the dependence on an integer parameter N is such that ρ(λ;N + 1) = λρ(λ;N). In this case the relevant quantities to consider (related to the Hankel determinants) are the coefficients of the three terms recurrence relation satisfied by these polynomials. The authors there proved that these quantities solve (up to some change of variables) the discrete- time Toda molecule equation, a coupled system of discrete equations in the two variables n, N . The result deeply relies on the quasi-periodic condition satisfied by the weight ρ. Back to our setting, the measure we have for our orthogonal polynomials on the unit circle is such that dµ(λ;N + 1) = e ∑N+1 j=1 θj j (eiλj+e−iλj)dλ 2π = e θN+1 N+1 (eiλ(N+1)+e−iλ(N+1)) dµ(λ;N). This relation does not seem as promising as the one for ρ for the study of the N -dependence, but it is another point that we could further investigate. Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 27 A The continuous limit This appendix contains further computations for the continuous limit of the equations of the discrete Painlevé II hierarchy (1.10) in the first cases N = 1, 2, 3. To obtain it, we follow the scaling limit given in [5, Theorem 1] as already recalled in the introduction. The case N = 1. Notice that in this case we recover the same computation done in [6, Chapter 9]. We consider equation (1.13) written as xn+1 + xn−1 + nxn θ1 ( 1− x2n ) = 0 in which the only parameter appearing is θ1 = θ. Following the scaling limit of [5, Theorem 1], in the case N = 1, we have b = 2, d = 1 and xn = (−1)nθ− 1 3u(t) with t = (n− 2θ)θ− 1 3 . Now, for θ → +∞, we compute xn±1 ∼ (−1)n+1θ− 1 3u ( t± θ− 1 3 ) ∼ (−1)n+1θ− 1 3 ( u(t)± θ− 1 3u′(t) + θ− 2 3 2 u′′(t) +O ( θ−1 )) , that gives xn+1 + xn−1 ∼ (−1)n+12θ− 1 3u(t) + (−1)n+1θ−1u′′(t) +O ( θ−1 ) . The other term appearing in the discrete Painlevé II equation gives instead nxn θ1 ( 1− x2n ) ∼ ( 2θ + tθ 1 3 ) (−1)nθ− 1 3u(t)θ−1 ( 1 + θ− 2 3u2(t) +O ( θ−1 )) ∼ (−1)n2θ− 1 3u(t) + (−1)nθ−1 ( tu(t) + 2u3(t) ) +O ( θ−1 ) . Thus equation (1.8) in this scaling limit gives at the first order (coefficient of θ−1) the second order differential equation u′′(t)− tu(t)− 2u3(t) = 0, which coincides indeed with the Painlevé II equation. The case N = 2. We consider equation (1.14), with the parameters θ1, θ2 rescaled as θ1 = θ, θ2 = θ 4 . It reads as nxn( 1− x2n ) + θ(xn+1 + xn−1) + θ 4 ( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 ) − xn(xn+1 + xn−1) 2 ) = 0 (A.1) and this time we consider the following scaling limit (case N = 2 in [5, Theorem 1]) b = 3 2 , d = 4 and xn = (−1)nθ− 1 5 4 1 5u(t) with t = ( n− 3 2 θ ) θ− 1 5 4 1 5 . For θ → +∞, similar computations gives the fourth order differential equation tu(t) + 6u(t)5 − 10u(t)u′(t)2 − 10u(t)2u′′(t) + u′′′′(t) = 0 28 T. Chouteau and S. Tarricone which corresponds to the second equation of the Painlevé II hierarchy. Detailed computations to obtain certain terms from the previous equation are given below. We begin with the expansion of the first term in equation (A.1): nxn( 1− x2n ) ∼ ( 3 2 θ + 4− 1 5 θ 1 5 t ) (−1)nθ− 1 5 4 1 5u(t) ( 1 + 4 2 5 θ− 2 5u2(t) + 4 4 5 θ− 4 5u4(t) +O ( θ−1 )) ∼ (−1)n ( 3 2 4 1 5 θ 4 5u(t) + 3 2 4 3 5 θ 2 5u(t)3 + tu(t) + 6u(t)5 +O ( θ− 1 5 )) . Computing expansions of xn±1, xn±2 as θ → ∞, we obtain xn±1 ∼ (−1)n+14 1 5 θ− 1 5u ( t± 4 1 5 θ− 1 5 ) ∼ (−1)n+14 1 5 θ− 1 5 × ( u(t)± 4 1 5 θ− 1 5u′(t) + 4 2 5 θ− 2 5 2 u′′(t)± 4 3 5 θ− 3 5 6 u′′′(t) + 4 4 5 θ− 4 5 24 u′′′′(t) +O ( θ−1 )) , xn±2 ∼ (−1)n4 1 5 θ− 1 5u ( t± 2θ− 1 5 4 1 5 ) ∼ (−1)n4 1 5 θ− 1 5 × ( u(t)± 4 1 5 2θ− 1 5u′(t) + 4 7 5 θ− 2 5u′′(t)± 4 8 5 2θ− 3 5 3 u′′′(t) + 4 9 5 θ− 4 5 3 u′′′′(t) +O ( θ−1 )) that gives for the second term of equation (A.1) θ(xn+1 + xn−1) ∼ (−1)n+1 ( 4 1 5 2θ 4 5u(t) + 4 3 5 θ 2 5u′′(t) + 1 3 u′′′′(t) +O ( θ− 1 5 )) . Some linear and nonlinear terms appear with the expansion of the third term of equation (A.1). The linear one is θ 4 (xn+2 + xn−2) ∼ (−1)n ( 4 1 5 θ 4 5 1 2 u(t) + 4 3 5 θ 2 5u′′(t) + 4 3 u′′′′(t) +O ( θ− 1 5 )) . Nonlinear ones are θ 4 xn(xn+1 + xn−1) 2 ∼ (−1)nu(t) ( 4 3 5 θ 2 5u(t)2 + 4u(t)u′′(t) +O ( θ− 1 5 )) , θ 4 xn±2x 2 n±1 ∼ (−1)n ( 4− 2 5 θ 2 5u(t)3 ± 4 4 5 θ 1 5u(t)2u′(t) + 3u(t)2u′′(t) + 5u(t)u′(t)2 ) . From these computations, we see that we recover exactly tu(t) + 6u(t)5 − 10u(t)u′(t)2 − 10u(t)2u′′(t) + u′′′′(t) = 0. The case N = 3. We consider equation (1.15) with the parameters θ1, θ2, θ3 rescaled as θ1 = θ, θ2 = 2θ 5 , θ3 = θ 15 and rewritten as nxn θ ( 1− x2n ) + (xn+1 + xn−1) + 2 5 ( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 ) − xn(xn+1 + xn−1) 2 ) + 1 15 ( x2n(xn+1 + xn−1) 3 + xn+3 ( 1− x2n+2 )( 1− x2n+1 ) + xn−3 ( 1− x2n−2 )( 1− x2n−1 )) + 1 15 ( −2xn(xn+1 + xn−1) ( xn+2 ( 1− x2n+1 ) + xn−2 ( 1− x2n−1 )) − xn−1x 2 n−2 ( 1− x2n−1 )) + 1 15 ( −xn+1x 2 n+2 ( 1− x2n+1 ) − xn+1xn−1(xn+1 + xn−1) ) = 0. Finally, we consider the following scaling limit (case N = 3 of [5, Theorem 1]) b = 4 3 , d = 15 and xn = (−1)nθ− 1 7 15 1 7u(t) with t = ( n− 4 3 θ ) θ− 1 7 15 1 7 . Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy 29 Again, for θ → +∞ the asymptotic expansion of the equation above results at the first order (coefficient of θ−1) into the sixth order differential equation tu(t) + 20u(t)7 − 140u(t)3u′(t)2 − 70u(t)4u′′(t) + 70u′(t)2u′′(t) + 42u(t)u′′(t)2 + 56u(t)u′(t)u′′′(t) + 14u(t)4u′′(t)− u′′′′′′(t) = 0, which corresponds to the third equation in the Painlevé II hierarchy. Remark A.1. Computations for N = 2 and N = 3 were performed with Maple/Mathematica. Files are available on demand. Acknowledgments We acknowledge the support of the H2020-MSCA-RISE-2017 PROJECT No. 778010 IPaDE- GAN and the International Research Project PIICQ, funded by CNRS. During the period from November 2021 to October 2022, S.T. was supported also by the Fonds de la Recherche Scientifique-FNRS under EOS project O013018F and based at the Institut de Recherche en Mathématique et Physique of UCLouvain. The authors are grateful to Mattia Cafasso for the inspiration given to work on this project and his guidance. The authors also want to thank the referees of this paper for useful comments and suggestions. S.T. is also grateful to Giulio Ruzza for meaningful conversations. References [1] Adler M., van Moerbeke P., Recursion relations for unitary integrals, combinatorics and the Toeplitz lattice, Comm. Math. Phys. 237 (2003), 397–440, arXiv:math-ph/0201063. [2] Baik J., Riemann–Hilbert problems for last passage percolation, in Recent Developments in Integrable Systems and Riemann–Hilbert Problems (Birmingham, AL, 2000), Contemp. Math., Vol. 326, Amer. Math. Soc., Providence, RI, 2003, 1–21, arXiv:math.PR/0107079. [3] Baik J., Deift P., Johansson K., On the distribution of the length of the longest increasing subsequence of random permutations, J. Amer. Math. Soc. 12 (1999), 1119–1178, arXiv:math.CO/9810105. [4] Baik J., Deift P., Suidan T., Combinatorics and random matrix theory, Grad. Stud. Math., Vol. 172, Amer. Math. Soc., Providence, RI, 2016. [5] Betea D., Bouttier J., Walsh H., Multicritical random partitions, Sém. Lothar. Combin. 85 B (2021), 33, 12 pages, arXiv:2012.01995. [6] Borodin A., Discrete gap probabilities and discrete Painlevé equations, Duke Math. J. 117 (2003), 489–542, arXiv:math-ph/0111008. [7] Borodin A., Okounkov A., A Fredholm determinant formula for Toeplitz determinants, Integral Equations Operator Theory 37 (2000), 386–396, arXiv:math.CA/9907165. [8] Cafasso M., Claeys T., Girotti M., Fredholm determinant solutions of the Painlevé II hierarchy and gap prob- abilities of determinantal point processes, Int. Math. Res. Not. 2021 (2021), 2437–2478, arXiv:1902.05595. [9] Cafasso M., Ruzza G., Integrable equations associated with the finite-temperature deformation of the discrete Bessel point proces, J. Lond. Math. Soc., to appear, arXiv:2207.01421. [10] Clarkson P.A., Joshi N., Mazzocco M., The Lax pair for the mKdV hierarchy, in Théories Asymptotiques et Équations de Painlevé, Sémin. Congr., Vol. 14, Soc. Math. France, Paris, 2006, 53–64. [11] Cresswell C., Joshi N., The discrete first, second and thirty-fourth Painlevé hierarchies, J. Phys. A 32 (1999), 655–669. [12] Dattoli G., Chiccoli C., Lorenzutta S., Maino G., Richetta M., Torre A., Generating functions of multivari- able generalized Bessel functions and Jacobi-elliptic functions, J. Math. Phys. 33 (1992), 25–36. [13] Deift P.A., Orthogonal polynomials and random matrices: a Riemann–Hilbert approach, Courant Lect. Notes Math., Vol. 3, Amer. Math. Soc., Providence, RI, 1999. https://doi.org/10.1007/s00220-003-0818-4 https://arxiv.org/abs/math-ph/0201063 https://doi.org/10.1090/conm/326/05753 https://doi.org/10.1090/conm/326/05753 https://arxiv.org/abs/math.PR/0107079 https://doi.org/10.1090/S0894-0347-99-00307-0 https://arxiv.org/abs/math.CO/9810105 https://doi.org/10.1090/gsm/172 https://doi.org/10.1090/gsm/172 https://arxiv.org/abs/2012.01995 https://doi.org/10.1215/S0012-7094-03-11734-2 https://arxiv.org/abs/math-ph/0111008 https://doi.org/10.1007/BF01192827 https://doi.org/10.1007/BF01192827 https://arxiv.org/abs/math.CA/9907165 https://doi.org/10.1093/imrn/rnz168 https://arxiv.org/abs/1902.05595 https://doi.org/10.1112/jlms.12745 https://arxiv.org/abs/2207.01421 https://doi.org/10.1088/0305-4470/32/4/009 https://doi.org/10.1063/1.529959 30 T. Chouteau and S. Tarricone [14] Flaschka H., Newell A.C., Monodromy- and spectrum-preserving deformations. I, Comm. Math. Phys. 76 (1980), 65–116. [15] Fokas A.S., Its A.R., Kitaev A.V., Discrete Painlevé equations and their appearance in quantum gravity, Comm. Math. Phys. 142 (1991), 313–344. [16] Forrester P.J., Witte N.S., Bi-orthogonal polynomials on the unit circle, regular semi-classical weights and integrable systems, Constr. Approx. 24 (2006), 201–237, arXiv:math.CA/0412394. [17] Hastings S.P., McLeod J.B., A boundary value problem associated with the second Painlevé transcendent and the Korteweg–de Vries equation, Arch. Rational Mech. Anal. 73 (1980), 31–51. [18] Hisakado M., Unitary matrix models and Painlevé III, Modern Phys. Lett. A 11 (1996), 3001–3010, arXiv:hep-th/9609214. [19] Hisakado M., Wadati M., Matrix models of two-dimensional gravity and discrete Toda theory, Modern Phys. Lett. A 11 (1996), 1797–1806, arXiv:hep-th/9605175. [20] Its A.R., Kitaev A.V., Fokas A.S., An isomonodromy approach to the theory of two-dimensional quantum gravity, Russian Math. Surveys 45 (1990), 155–157. [21] Kimura T., Zahabi A., Universal edge scaling in random partitions, Lett. Math. Phys. 111 (2021), 48, 16 pages, arXiv:2012.06424. [22] Le Doussal P., Majumdar S.N., Schehr G., Multicritical edge statistics for the momenta of fermions in nonharmonic traps, Phys. Rev. Lett. 121 (2018), 030603, 7 pages, arXiv:1802.06436. [23] Okounkov A., Infinite wedge and random partitions, Selecta Math. (N.S.) 7 (2001), 57–81, arXiv:math.RT/9907127. [24] Painlevé P., Mémoire sur les équations différentielles dont l’intégrale générale est uniforme, Bull. Soc. Math. France 28 (1900), 201–261. [25] Periwal V., Shevitz D., Exactly solvable unitary matrix models: multicritical potentials and correlations, Nuclear Phys. B 344 (1990), 731–746. [26] Ramani A., Grammaticos B., Hietarinta J., Discrete versions of the Painlevé equations, Phys. Rev. Lett. 67 (1991), 1829–1832. [27] Schensted C., Longest increasing and decreasing subsequences, Canadian J. Math. 13 (1961), 179–191. [28] Tracy C.A., Widom H., Fredholm determinants, differential equations and matrix models, Comm. Math. Phys. 163 (1994), 33–72, arXiv:hep-th/9306042. [29] Tracy C.A., Widom H., Level-spacing distributions and the Airy kernel, Comm. Math. Phys. 159 (1994), 151–174, arXiv:hep-th/9211141. https://doi.org/10.1007/BF01197110 https://doi.org/10.1007/BF02102066 https://doi.org/10.1007/s00365-005-0616-7 https://arxiv.org/abs/math.CA/0412394 https://doi.org/10.1007/BF00283254 https://doi.org/10.1142/S0217732396002976 https://arxiv.org/abs/hep-th/9609214 https://doi.org/10.1142/S0217732396001788 https://doi.org/10.1142/S0217732396001788 https://arxiv.org/abs/hep-th/9605175 https://doi.org/10.1070/RM1990v045n06ABEH002699 https://doi.org/10.1007/s11005-021-01389-y https://arxiv.org/abs/2012.06424 https://doi.org/10.1103/PhysRevLett.121.030603 https://arxiv.org/abs/1802.06436 https://doi.org/10.1007/PL00001398 https://arxiv.org/abs/math.RT/9907127 https://doi.org/10.24033/bsmf.633 https://doi.org/10.24033/bsmf.633 https://doi.org/10.1016/0550-3213(90)90676-5 https://doi.org/10.1103/PhysRevLett.67.1829 https://doi.org/10.4153/CJM-1961-015-3 https://doi.org/10.1007/BF02101734 https://doi.org/10.1007/BF02101734 https://arxiv.org/abs/hep-th/9306042 https://doi.org/10.1007/BF02100489 https://arxiv.org/abs/hep-th/9211141 1 Introduction 2 OPUC: the Riemann–Hilbert approach and a discrete Painlevé II Lax pair 2.1 Toeplitz determinants related to OPUC 2.2 Riemann–Hilbert problem associated to OPUC 2.3 A linear differential system for Psi(z) 2.4 Relation with the Cresswell–Joshi Lax pair 3 From the Lax Pair to the discrete Painlevé II hierarchy 3.1 The symmetry in the compatibility condition 3.2 The recursion 3.3 The relation between T_{i,12}(n) and T_{i,21}(n) A The continuous limit References
id nasplib_isofts_kiev_ua-123456789-211913
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
issn 1815-0659
language English
last_indexed 2026-03-17T00:46:36Z
publishDate 2023
publisher Інститут математики НАН України
record_format dspace
spelling Chouteau, Thomas
Tarricone, Sofia
2026-01-16T11:18:15Z
2023
Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy. Thomas Chouteau and Sofia Tarricone. SIGMA 19 (2023), 030, 30 pages
1815-0659
2020 Mathematics Subject Classification: 33E17; 33C47; 35Q15
arXiv:2211.16898
https://nasplib.isofts.kiev.ua/handle/123456789/211913
https://doi.org/10.3842/SIGMA.2023.030
Solutions of the discrete Painlevé II hierarchy are shown to be in relation to a family of Toeplitz determinants describing certain quantities in multicritical random partition models, for which the limiting behavior has been recently considered in the literature. Our proof is based on the Riemann-Hilbert approach for the orthogonal polynomials on the unit circle related to the Toeplitz determinants of interest. This technique allows us to construct a new Lax pair for the discrete Painlevé II hierarchy that is then mapped to the one introduced by Cresswell and Joshi.
We acknowledge the support of the H2020-MSCA-RISE-2017 PROJECT No. 778010 IPaDEGAN and the International Research Project PIICQ, funded by CNRS. During the period from November 2021 to October 2022, S.T. was also supported by the Fonds de la Recherche Scientifique-FNRS under the EOS project O013018F and based at the Institut de Recherche en Mathématique et Physique of UCLouvain. The authors are grateful to Mattia Cafasso for the inspiration given to work on this project and his guidance. The authors also want to thank the referees of this paper for their useful comments and suggestions. S.T. is also grateful to Giulio Ruzza for meaningful conversations.
en
Інститут математики НАН України
Symmetry, Integrability and Geometry: Methods and Applications
Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
Article
published earlier
spellingShingle Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
Chouteau, Thomas
Tarricone, Sofia
title Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
title_full Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
title_fullStr Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
title_full_unstemmed Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
title_short Recursion Relation for Toeplitz Determinants and the Discrete Painlevé II Hierarchy
title_sort recursion relation for toeplitz determinants and the discrete painlevé ii hierarchy
url https://nasplib.isofts.kiev.ua/handle/123456789/211913
work_keys_str_mv AT chouteauthomas recursionrelationfortoeplitzdeterminantsandthediscretepainleveiihierarchy
AT tarriconesofia recursionrelationfortoeplitzdeterminantsandthediscretepainleveiihierarchy