- Research
- Open access
- Published:
Jordan decomposition and geometric multiplicity for a class of non-symmetric Ornstein-Uhlenbeck operators
Advances in Difference Equations volume 2014, Article number: 34 (2014)
Abstract
In this paper, we calculate the Jordan decomposition for a class of non-symmetric Ornstein-Uhlenbeck operators with the drift coefficient matrix, being a Jordan block, and the diffusion coefficient matrix, being the identity multiplying a constant. For the 2-dimensional case, we present all the general eigenfunctions by mathematical induction. For the 3-dimensional case, we divide the calculation of the Jordan decomposition into three steps. The key step is to do the canonical projection onto the homogeneous Hermite polynomials, and then use the theory of systems of linear equations. Finally, we get the geometric multiplicity of the eigenvalue of the Ornstein-Uhlenbeck operator.
1 Introduction
For the symmetric Ornstein-Uhlenbeck operator, the eigenfunctions are the well-known Hermite polynomials [1]. The eigenfunctions of a type of finite-dimensional normal but non-symmetric Ornstein-Uhlenbeck operators have recently been found. They are the so-called complex Hermite polynomials [2] (or, say, the Hermite-Laguerre-Itô polynomials) where the idea is to proceed by means of a decomposition to the summation of series of up to a 2-dimensional normal Ornstein-Uhlenbeck operator [3]. But if the Ornstein-Uhlenbeck operator is not normal, the general eigenfunctions are still unknown up to now.
In the present paper, we consider the d-dimensional () non-symmetric Ornstein-Uhlenbeck process
The associated Ornstein-Uhlenbeck operator is
Denote with Id the identity and R the nilpotent. Clearly, is a non-symmetric operator since B does not satisfy the reversiblea condition of Ornstein-Uhlenbeck [4]
where is the diffusion coefficient matrix and is the transpose matrix of B.
The associated Markov semigroup on the Banach space of the bounded measurable functions is
and
It is well known that extends to a strongly continuous semigroup of positive contractions in the Hilbert space , where μ is the unique invariant measure [5, 6]. Still we denote by the generator of in ; it was shown [7] that the spectrum consists of eigenvalues of finite multiplicities, , and all the generalized eigenfunctions are polynomials and form a complete system in . Let . It follows from [[7], Theorem 4.1] that the algebraic multiplicity of γ is
and it follows from [[7], Proposition 4.3, Theorem 4.1] that , the index of the eigenvalue γ, is
A natural question is what the geometric multiplicity of the eigenvalue γ is. In addition, since the spectral subspace associated with γ (i.e. ) is a finite-dimensional vector space over the real field ℝ, what is the Jordan decomposition (or the Jordan canonical form) [8, 9] of restricted on the spectral subspace? That is to say, what are the integers , and the generalized eigenfunctions , such that
form the basis of the spectral subspace associated to γ, and
The integers are also called Segre characteristic (or, say, Segre type, Segre notation). is called a lead vector (or, say, a cyclic vector, a generator) of a Jordan chain by some authors [9, 10].
In the present paper, we present an approach to calculate the Jordan decomposition and the generalized eigenfunctions (see Theorems 2.1, 3.1) for .b The proof of Theorem 2.1 is by direct calculation. The main techniques of the proof of Theorem 3.1 are canonical projection and the theory of systems of linear equations. This approach is novel to the Jordan decomposition of differential operators as far as we know.
It is a difficult problem to get the geometric multiplicity of the eigenvalue of a differential operator from the perspective of functional analysis. It is well known that the spectrum of the symmetric Ornstein-Uhlenbeck operator is the starting point of stochastic analysis (more precisely: Malliavin calculus), thus the analogous results of the non-symmetric Ornstein-Uhlenbeck operator are interesting. We will treat the more general non-symmetric Ornstein-Uhlenbeck operator in the future.
2 In case of dimension 2
In this section, we treat the case of . Denote . The Hermite polynomials [1] are defined by the formula
Clearly, it has a power series expression,
and satisfies
Theorem 2.1 The geometric multiplicity of the eigenvalue γ is 1. Set
Suppose , then forms a basis of the spectral subspace associated to γ. Also
where and when . In particular, is the eigenfunction associated to γ.
Proof We need only prove Eq. (2.2). It is easy to check that satisfies the recursion relation:
Clearly, for any differentiable function , , we have
Then by the property of the Hermite polynomials [1], we have
By mathematical induction, we have
□
3 In case of dimension 3
In this section, we treat the case of . For convenience, we give some notation firstly. Set for the space of all polynomials with variables , the space of polynomials of degree less than or equal to n, and the space of homogeneous polynomials of degree n. Then and one has the usual direct sum decomposition of all polynomials [7],
Notation 1 Set , and .
By the monomials property of the Hermite polynomials [1]
the Hermite polynomials are another basis of .
Let , then we have another direct sum decomposition of all polynomials, i.e.,
We denote by the canonical projection [11] of onto .
Theorem 3.1 Let
Then there exist so that
forms a basis of the spectral subspace associated to γ. Set . Then are the basis of the eigenspace of the eigenvalue γ and satisfy
The proof of Theorem 3.1 is presented in Section 3.1. The following is a by-product.
Corollary 3.2 The geometric multiplicity of the eigenvalue γ of the Ornstein-Uhlenbeck operator is .
3.1 Proof of Theorem 3.1
Note that
It follows from the property of the Hermite polynomials [1] that
For convenience, Eq. (3.5) can be rewritten in the following way.
Proposition 3.3 If then . In particular,
where and
and
In particular, if , then
Remark 1 Equations (3.7)-(3.9) make us use the terminology of graph theory. In fact, by Eq. (3.9), we have a weighted and directed acyclic graph (which also can be seen as a Hasse diagram) of the evolution of basis of operated by . For example, when and ,c the directed acyclic graphs are, respectively,
where we denote by the triple integers the Hermite polynomial . It follows from Eq. (3.9) that the weights of the arrows between and , are −j, −i, respectively.
One can find many properties from the directed acyclic graph. For example, for the vertex with , the height (which is defined by the distance between vertices and and thus is between ) is . For simplicity, in each height of the graph, we list the vertices decreasingly by lexicographic order. Then the vertices and are symmetric about the n th height of the graph.
It follows from Eq. (1.4), Corollary 3.2, and Theorem 3.1 that the order of the graph (the numbers of vertices in the graph) is the algebraic multiplicity of γ, are 1 plus the distance between vertices and , and the numbers of vertices in the n th height of the graph is the geometric multiplicity of γ.
Proposition 3.4 The spectral subspace associated to belongs to .
Proof Suppose that f is a generalized eigenfunction, i.e., there exists an integer such that . It follows from Eq. (3.7) that if the degree of f is , then . This is a contradiction; then and the degree is exactly n.d If there is an such that , then by Eq. (3.6), . This is a contradiction; then . □
Lemma 3.5 For any polynomial with , there exists a unique solution to the equation .
Proof Suppose that
and . It follows from Eq. (3.7) that
By the linear independence of , we have a system of linear equations in unknowns.
We sort , which appear in the directed acyclic graph in Remark 1. The coefficient matrix of the linear equations is a lower triangle matrix with nonzero diagonal entry . Thus the linear equations have a unique solution. □
Suppose that with . It follows from Eq. (3.6) that
Thus the equation is equal to the system of equations
It follows from Lemma 3.5 that when , Eq. (3.10) has a unique solution and when , Eq. (3.11) has a unique solution .
Clearly, if f satisfies , so does where h is any eigenfunction of associated to γ. Thus we have the following proposition.
Proposition 3.6 For any with , , there exist solutions to the equation . In addition, if f is the same degree polynomial to g then there exists one and only one solution.
Proposition 3.7 Set , . Then it satisfies the equation .
Proof Suppose that . By Eq. (3.7), we have
By the linear independence of , we have a system of k linear homogeneous equations in unknowns. The coefficient matrix is
Clearly, the solution is 1-dimensional and , , is a solution. □
Now suppose that with and ; then by Eq. (3.6),
Therefore, the equation is equivalent to two equations:
By Proposition 3.7, Eq. (3.12) has independent solutions. It follows from Proposition 3.6 that there exists a unique ϕ satisfying Eq. (3.13) for each ψ. Thus we have the following corollary.
Corollary 3.8 The geometric multiplicity of the eigenvalue is greater than or equal to (i.e., there are at least independent solutions to the equation ).
Denote by , the solutions to the equation given by Corollary 3.8. Clearly, , where is the same as in Proposition 3.7.
Suppose that with and , then
Therefore, the equation is equivalent to two equations:
Note that . Set , then the l.h.s. of Eq. (3.14) is a linear mapping from the to the . The linear mapping is the evolution from the 2k-height to the -height of the directed acyclic graph in Remark 1, which is represented in the natural basis by a -square matrix (it is the multiplication of some matrices; for details, please refer to Section 3.2). By Proposition 3.10, the matrix is nonsingular, which implies that Eq. (3.14) has a solution. Since , it follows from Proposition 3.6 that Eq. (3.15) has a solution. Then we have the following proposition.
Proposition 3.9 There exists an such that .
Proof of Theorem 3.1 Note that in Proposition 3.7 are linear independent, so are the eigenfunctions . Let be as in Proposition 3.9. Then the generalized eigenfunctions are linear independent (please refer to the proof of [[12], p.264, Theorem 6.2.]).
Note that ; then the algebraic multiplicity of the eigenvalue γ can be decomposed to
Thus forms a basis of the spectral subspace associated to γ. Together with Proposition 3.8, we see that the geometric multiplicity of the eigenvalue γ is equal to (otherwise, the algebraic multiplicity should be greater than ). Then forms the basis of the eigenspace of γ. Equation (3.4) is exactly the conclusion of Proposition 3.7. □
3.2 Linear mapping represented by the multiplication of some matrices
For example, by Eq. (3.9), when n is odd, the evolution from the -height to the n-height (the n-height to the -height) of the directed acyclic graph is
Then the matrices associated to the linear mappings are , (see below). When n is even, the evolution from the -height to the n-height (the n-height to the -height) of the directed acyclic graph is
Then the matrices associated to the linear mappings are , (see below).
The others are similar. In general, the matrix (see Proposition 3.9) associated to the linear mapping of the evolution from the -height to the -height of the directed acyclic graph is as follows.
Proposition 3.10 Let . If n is odd, then suppose that is an -square matrix given by
where , are -square matrices,
and
and , are , matrices, respectively,
and
If n is even, then suppose that is an -square matrix given by
where , are order matrices,
and
and , are , matrices, respectively,
and
Then the matrices are nonsingular.
Remark 2 Set the column vector , where . We conjecture that the column vector is the eigenvector of associated to the eigenvalue , which is defined as follows: when n is odd,
when n is even,
If the conjecture is valid, then we can characterize the leader vectors more clearly, i.e.,
Apply the notation in [10, 13]. Let A be an matrix, , and , , . Denote by , or simply , the submatrix of A consisting of the entries in rows and columns . Let be the submatrix of A obtained by deleting rows and columns . For convenience, () means deleting rows (columns) only.
Proof We divide the proof into three steps.
Claim 1: All the minors (i.e., the determinant of the square submatrix) of the matrices Ak, , , , , are nonnegative. Since () and () have the same types, we only need to prove the cases of , . The square submatrix of is the same as that of a certain , , which is a direct sum of two matrices [13] with the same type of or . Thus we only need to prove the case of . In fact, is a direct sum [13] of several nonnegative triangular matrices, which implies that has a nonnegative determinant.
Claim 2: All the minors of are nonnegative. In fact, we have the following type of Binet-Cauchy formula:
where A (B) is () matrix, , κ runs over all the sequences of with the same length of α, β. By successively applying Eq. (3.28), we have
Together with Claim 1 and by mathematical induction, we have .
Claim 3: . Clearly, . Since , we need only prove that . By the Binet-Cauchy formula, we have
Clearly, . By the induction assumption , Claim 2 implies that there exists at least one . This ends the proof. □
Endnotes
a It is well known that the reversibility is equivalent to the symmetric property of the generator for a Markov process with invariant distribution.
b We think that the approach is also able to solve the same problems for .
c The directed acyclic graph can easily be given for any integer n, and we ignore it for convenience.
d The reader can also refer to [[7], Proposition 3.1].
References
Kuo HH: Introduction to Stochastic Integration. Springer, Berlin; 2006.
Itô K: Multiple Wiener integral. J. Math. Soc. Jpn. 1951, 3(1):157-169. (Reprinted in: Kiyosi Itô selected papers. Edited by Daniel W. Stroock, S.R.S. Varadhan, Springer (1987)) 10.2969/jmsj/00310157
Chen, Y, Liu, Y: On the eigenfunctions of the complex Ornstein-Uhlenbeck operator. Kyoto J. Math. (2013, to appear)http://arxiv.org/pdf/1209.4990.pdf
Chojnowska-Michalik A, Goldys B: Symmetric Ornstein-Uhlenbeck semigroups and their generators. Probab. Theory Relat. Fields 2002, 124(4):459-486. 10.1007/s004400200222
Chojnowska-Michalik A, Goldys B: Nonsymmetric Ornstein-Uhlenbeck semigroup as second quantized operator. J. Math. Kyoto Univ. 1996, 36(3):481-498.
Lunardi A:On the Ornstein-Uhlenbeck operator in spaces with respect to invariant measures. Trans. Am. Math. Soc. 1997, 349: 155-169. 10.1090/S0002-9947-97-01802-3
Metafune G, Pallara D:Spectrum of Ornstein-Uhlenbeck operators in spaces with respect to invariant measures. J. Funct. Anal. 2002, 196: 40-60. 10.1006/jfan.2002.3978
Levelt AHM: Jordan decomposition for a class of singular differential operators. Ark. Mat. 1975, 13(1):1-27.
Mathew, PJ: On Some Aspects of the Differential Operator. Mathematics Theses (2006)
Hoffman K, Kunze R: Linear Algebra. 2nd edition. Prentice Hall, New York; 1971.
Roman S: Advanced Linear Algebra (GTM135). 3rd edition. Springer, Berlin; 2008.
Lang S: Linear Algebra. 3rd edition. Springer, Berlin; 2004.
Zhang FZ: Matrix Theory, Basic Results and Techniques. 2nd edition. Springer, Berlin; 2011.
Acknowledgements
This work was supported by NSFC (No. 11101137).
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors did not provide this information.
Authors’ contributions
The authors did not provide this information.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License ( https://creativecommons.org/licenses/by/2.0 ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Rao, Y., Wang, J. & Chen, Y. Jordan decomposition and geometric multiplicity for a class of non-symmetric Ornstein-Uhlenbeck operators. Adv Differ Equ 2014, 34 (2014). https://doi.org/10.1186/1687-1847-2014-34
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/1687-1847-2014-34