ABSTRACT
In this paper, we give a geometric construction for different allowable properties for sign pattern matrices. In Section 2, we give a construction for detecting a sign pattern matrix to be potentially nilpotent and also compute the nilpotent matrix realization for a given sign pattern matrix if it exists. In Section 3, we develop a geometric construction for potential stability. In Section 4, we establish a necessary and sufficient condition for a sign pattern matrix to be spectrally arbitrary. For a given sign pattern matrix of order , we prove that there exists a surface of dimension at most for such that for every vector on the same surface, there exists a matrix , a qualitative class of whose characteristic polynomial is .
1. Introduction
A sign pattern matrix of order is an matrix whose entries belong to the sign set . A qualitative class of a sign pattern matrix is denoted by and is defined as
where is the entry of a sign pattern matrix . A property of an matrix is said to be an allowable property for a given sign pattern matrix if there exists an matrix such that satisfies property . A sign pattern matrix requires a property if every matrix in satisfies property .
A permutation pattern is a square sign pattern matrix with entries and , where the entry occurs precisely once in each row and in each column. A signature sign pattern is a square sign pattern having diagonal entries either or and off-diagonal entries . Sign patterns and are said to be permutationally similar if there exists a permutation pattern matrix such that . Sign patterns and are said to be signature similar if there exists a signature sign pattern matrix such that . Two sign patterns are said to be equivalent if one of them can be obtained from other by using any combination of transpositions, permutation similarity and signature similarity. In 1947, an economist P. A. Samuelson considered special matrices while studying mathematical modeling of problems from economics. Entries of these matrices were signs instead of real numbers. Such matrices also arise in population biology, chemistry, sociology and many other situations. A study of these matrices falls under the branch of combinatorial matrix theory. In this paper, we have studied some of these qualitative matrix problems.
In Section 2, we have discussed about potentially nilpotent sign pattern matrices, computed nilpotent realization for a given sign pattern matrix if it exists, and also discussed about an index of nilpotency for computed nilpotent realizations. In Section 3, we have introduced the concept of arbitrary potentially stable sign pattern matrices, a method for detecting arbitrary potential stability of a sign pattern matrix. In the same section, we have characterized order 2 arbitrary potentially stable sign pattern matrices. In Section 4, spectrally arbitrary sign pattern matrices have been investigated. This section also generalizes some results from Section 2.
2. Potentially nilpotent sign pattern matrices
A non-zero matrix of order is said to be nilpotent if there exists a positive integer with . The smallest positive integer is called as an index of nilpotency for the matrix .
Definition 2.1 (Hogben et al., Citation2018). A sign pattern matrix of order is said to be potentially nilpotent if allows a nilpotent matrix, i.e., there exists a nilpotent matrix in the qualitative class of .
We will assign a one-to-one correspondence between vectors in and coefficients of a characteristic polynomial for a matrix in . For a vector , there is a characteristic polynomial for some square matrix of order .
Definition 2.2 (Jadhav & Deore, Citation2022). Let be a given sign pattern matrix of order . A matrix is said to be a realization for a given vector in if the characteristic polynomial of is
Here, we discuss about two open questions, the first question has been listed in Catral et al. (Citation2009) and Eschenbach and Johnson (Citation1988) and the second question which has not been directly appeared in the literature. Bergsma et al., Citation2012; Luo et al., (Citation2015) have discussed about potentially nilpotent sign pattern matrices. However, in Catral et al. (Citation2009), we observe that the nilpotent-Jacobian method for proving spectral arbitrariness of a given sign pattern matrix requires the explicit computation of the nilpotent matrix from the qualitative class of a given sign pattern matrix.
Question 1. Is a sign pattern matrix potentially nilpotent?
If we want to check whether a given sign pattern matrix is potentially nilpotent, it is sufficient to find a matrix realization for the vector .
Question 2. If a sign pattern is potentially nilpotent, then how do we find a nilpotent matrix in the qualitative class of ?
In Theorem 2.4, we prove the sufficient condition for sign pattern matrices to be potentially nilpotent. Throughout this article, the characteristic polynomial of a square matrix will be denoted by .
Lemma 2.3 (Jadhav & Deore, Citation2022). Let and be two matrices over the field of real numbers such that they vary either in a fixed row or in a fixed column. Then
Above lemma is easy to prove and can be extended to the convex linear combination of matrices.
Theorem 2.4. Let be a sign pattern matrix of order . Suppose and are two matrices in such that and vary exactly either in one row or in one column. If the vectors correspond to and are with same magnitude and in opposite direction of each other, then a sign pattern is potentially nilpotent.
Proof. Suppose that matrix realizations and vary exactly either in a row or a column. Let and be the characteristic polynomials of and , respectively, so that the vector corresponds to matrix and the vector corresponds to matrix . Now, by hypothesis, we have , i.e., . From Lemma 2.3, the characteristic polynomial of the matrix is given by . Since , we have . This shows that is nilpotent and . Thus, gives a required nilpotent matrix realization for a sign pattern . Hence, is potentially nilpotent.
A generalization of Theorem 2.4 is the following.
Theorem 2.5. Let be a sign pattern matrix of order . Suppose and are matrices in such that , vary exactly either in one row or in one column. If and are the vectors corresponding to and , respectively, and satisfy for some , then is a potentially nilpotent sign pattern matrix.
Proof. In view of Lemma 2.3, the characteristic polynomial of is . As and are realizations of vectors and such that they vary exactly in a row or a column, the matrix is a realization for the vector . Also, . Thus, is the required nilpotent matrix. Hence, the result.
Remark 1. Considering an appropriate convex linear combination, Theorem 2.5 holds for any number of matrices.
Example 2.1. Consider an example of a sign pattern matrix of order 5 as given below
Lemma 2.6. If nilpotent matrices and of order vary only in a row or in a column, then is nilpotent for all .
Proof. As and are nilpotent matrices of order , the characteristic polynomial of and is . Thus, Lemma 2.3 says that the characteristic polynomial of is for all . Hence, the matrix is nilpotent for every .
Lemma 2.7. Let and be square matrices that vary exactly either in a row or in a column and . Then, for all .
Proof. If and vary exactly in the column, then by splitting determinant of along the column, we get for all , hence the result.
Remark 2. For square matrices and , the determinant of is zero for all if determinant of and is zero with and varying exactly either in a row or in a column.
We denote the column of a matrix by for all values of in the proof of the following Lemma. Now, we analyze an index of nilpotent matrices obtained from the construction as given in Theorems 2.4 and 2.5.
Lemma 2.8. Let and be square matrices of order with rank such that they vary exactly in a row or in a column. Then, rank of is either for all or for some .
Proof. The rank of a matrix is preserved under permutation similarity. Without loss of generality, assume that matrices and vary in the last column. Let be the columns of and let be the columns of . As , we have the following two cases:
Case 1. Column does not belong to a set of linearly independent columns of . By rearranging the columns (if necessary), assume that the first columns of are linearly independent. Thus, remaining columns are linearly dependent on first columns. Hence, we have and . Therefore, the column of , i.e., where for all . Thus, in a matrix , the column is linearly dependent on its first columns, and the remaining columns are already linearly dependent on first columns, as they are the same as in matrices and . It follows that the rank of is .
Case 2. Column belongs to the set of linearly independent columns. In this case for all , if the column is linearly independent to remaining linearly independent columns, then rank of still remains as . Otherwise, for some , the column is linearly dependent on remaining linearly independent columns, and then the rank of is for these values of .
By the rank-nullity theorem, for a nilpotent matrix of rank , the dimension of its kernel space is ; it means that the dimension of the eigenspace corresponding to an eigenvalue is . Hence, the number of Jordan blocks corresponding to an eigenvalue , which is equal to the geometric multiplicity of an eigenvalue , has to be . Therefore, the largest possible size of a Jordan block is , and when we distribute over blocks almost equally, the minimal possible size of a larger Jordan block amongst them is (where denotes the smallest integer but not smaller than ). Now for any nilpotent matrix, its index is nothing but the size of a larger Jordan block in its Jordan canonical form. Hence, the index of nilpotency for a rank nilpotent matrix is at most and at least .
Theorem 2.9. Let be a sign pattern matrix of order . The nilpotent realization of obtained by Theorems 2.4 and 2.5 has index of nilpotency at most or and at least or for matrices and of rank .
Proof. Given that is a nilpotent matrix realization of a sign pattern matrix , computed by Theorem 2.4 or 2.5. Moreover, and are matrices of rank . By using Lemma 2.8, the nilpotent matrix has rank or . Therefore, as discussed in the above paragraph before Theorem 2.9, we conclude that the index of nilpotency for a matrix is at most or and at least or .
In example 2.1, we observe that the index of nilpotency for the matrix is .
3. Potentially stable sign pattern
An matrix is said to be a stable matrix if all of its eigenvalues have negative real parts. A sign pattern is said to be potentially stable if it allows stability, i.e., there exists a stable matrix in its qualitative class. Grundy et al. (Citation2012) discussed about constructions of potentially stable sign pattern matrices. Cavers (Citation2021) used polynomial stability to show that certain sign patterns are not potentially stable. In this section, we are introducing arbitrary potential stability of sign pattern matrices.
Let be an matrix with real entries. If all eigenvalues of have negative real parts, then all the coefficients of its characteristic polynomial are positive. Let and be stable matrices of order . Is a stable matrix?
Example 3.1. Let and. Then and are stable matrices, and they differ only in the first row, but is not stable.
But nevertheless, we have the following result true.
Theorem 3.1. Let be a sign pattern matrix. Suppose and are two stable matrices in the qualitative class such that they vary either in a row (or a column). Then, real eigenvalues of for are negative.
Proof. Without loss of generality, assume that matrices and vary in the row. Since and are stable matrices, all eigenvalues of and have negative real parts. Hence, all the coefficients of the characteristic polynomial of as well as the coefficients of the characteristic polynomial of are positive. Let and be characteristic polynomials of and , respectively, where all s and s are positive. Let . By Lemma 2.3, the characteristic polynomial
As all s and s are positive, we have is positive for all . If a real number is an eigenvalue of , then from EquationEquation 1(1) (1) , we have
which is not possible as the left hand side of the above equation is non-negative, but the right hand side is strictly negative. Hence, every real eigenvalue of has to be negative.
Remark 3. Theorem 3.1 can also be extended to matrices corresponding to unit vectors surrounding a hyperoctant in .
We give here the sufficient condition for potential stability of a sign pattern matrix of order . Note that the proof of the following theorem is essentially as similar to the proof of Theorem 2.7 given in Jadhav and Deore (Citation2022). For the sake of completeness, we have incorporated the same.
Theorem 3.2. Let be a sign pattern matrix of order . Suppose there exist matrices in , which are realizations of the vectors respectively. Further, assume that all these matrices vary exactly either in a row or in a column. Then the sign pattern is potentially stable.
Proof. We shall prove that every point lying in the hyperoctant surrounded by the vectors (denote it by ) is realized by a matrix in . Let be any point lying in the hyperoctant . Consider the curve for all and part of the plane lying in the hyperoctant . Note that the plane and the curve intersect exactly at one point say . Assume that the point corresponds to on the curve. Also, note that . As are realizations for the vectors , in , by Lemma 2.3, every point on the convex linear combination of has a matrix realization lying in . Thus, has a matrix realization say in . But then lying in , provides a matrix realization for the point . So every point in the hyperoctant has a matrix realization in .
In particular, the point lies in the hyperoctant ; hence, it has a matrix realization in . Also, polynomial corresponding to this point is ; this proves is potentially stable sign pattern.
Let us illustrate the above theorem with the help of an example.
Example 3.2. Consider a sign pattern matrix of order as given below:
Consider a realization in a qualitative class of where are positive real numbers. Its characteristic polynomial is given as follows
Now to find the values of and so that the polynomial in 2 corresponds to a vector , i.e, it becomes . Equating with , we get a system of linear equations in variables and . Solving this system of equations, we get . Therefore, we get a matrix in the qualitative class of whose characteristic polynomial is . Similarly, we obtained the matrices
, , and having the characteristic polynomials and , respectively. Observe that all the matrices have the same sign pattern and they vary only in the first row. Also they are realizations of vectors , respectively. Hence, by Theorem 3.2, a sign pattern matrix given by is potentially stable.
Remark 4. Theorem 3.2 says something extra rather than only saying potential stability of a sign pattern matrix . Potential stability means that a sign pattern allows a stable matrix. But for sign pattern matrices whose potential stability is proved by Theorem 3.2, we have for any size multiset of complex numbers with real parts negative and closed under complex conjugation, there exists a matrix in whose set of eigenvalues is the given multiset. Therefore, a sign pattern allows all possible stable matrices which have the sign pattern as that of .
Definition 3.3. A square sign pattern matrix is said to be an arbitrary potentially stable sign pattern if for every multiset of complex numbers with real parts negative and closed under complex conjugation, there exists a matrix in whose set of eigenvalues is the given multiset.
A sign pattern given in Example 3.2 is arbitrary potentially stable. A potentially stable sign pattern matrix is not need to be arbitrary potentially stable, and we can see the same in the following example.
Example 3.3. Consider a sign pattern given by . From Catral et al. (Citation2009), we note that allows a stable matrix specifically , so it is potentially stable. Consider a realization of obtained by replacing all non-zero entries by variables say where are all positive real numbers. The matrix corresponds to a vector . By equating vector with , we get and . Substituting back the value of , we observe that and , which is not possible. Thus, a vector can never be equal to . It means a vector can never be realized by a matrix in the qualitative class of sign pattern . Thus, a polynomial can never be a characteristic polynomial of any matrix in the qualitative class of . Thus, is not arbitrary potentially stable.
However, Theorem 3.2 gives only a sufficient condition for an arbitrary potentially stable sign pattern. It is not a necessary condition, observed from the following example of order sign pattern matrix.
Example 3.4. Consider a sign pattern matrix .Let be a matrix realization of sign pattern , where are all positive real numbers. This matrix corresponds to a vector . It can be observed that for any vector lying in the second quadrant , there exist values of with and . It means that a sign pattern is an arbitrary potentially stable sign pattern matrix. If we equate vector with , we get and . As being all are positive real numbers, we cannot have a solution to . Similarly, vector cannot be equal to , so that Theorem 3.2 is not applicable but still sign pattern is arbitrary potentially stable.
Now, we will give the characterizations of arbitrary potentially stable sign pattern matrices. It should be noted that every spectrally arbitrary sign patterns (Definition 4.1) are always arbitrary potential stable.
Theorem 3.1. A sign pattern matrix is arbitrary potentially stable if it is a transposition or a permutation similarity equivalent to any of the following sign patterns
Proof. A sign pattern matrix is arbitrary potentially stable as being spectrally arbitrary. Working out as in Example 3.4, we get sign pattern matrices are arbitrary potentially stable. Now, sign pattern matrices containing either four zero or three zero entries cannot be potentially stable as being every matrix in their qualitative class has zero determinant. Similarly, sign pattern matrices having two zero entries in the same row or in the same column cannot be potentially stable as being zero determinant. Sign pattern matrices having two zero entries on the diagonal cannot be potentially stable as being every matrix in their qualitative class has trace zero. Similarly, sign pattern matrices and , where denotes either or , cannot be potentially stable, as matrices in their qualitative class have either positive trace or negative determinant. Let be any matrix in the qualitative class of a sign pattern matrix , where are positive real numbers. It has a corresponding vector in . If we equate , then there is no solution with and in positive real numbers. However, the roots of the polynomial have negative real parts. Thus, a sign pattern matrix is not arbitrary potentially stable. Similarly, it can be proved that the remaining order sign pattern matrices cannot be arbitrary potentially stable.
It should be noted that an arbitrary potential stability is not preserved under signature similarity, as explained in the following example.
Example 3.2. Consider an arbitrary potentially stable sign pattern matrix and a signature matrix . Then, , which is not arbitrary potentially stable because every matrix in the qualitative class of has positive trace.
4. Spectrally arbitrary sign pattern matrices
Definition 4.1 (Catral et al., Citation2009). A sign pattern matrix of order is said to be spectrally arbitrary if every monic polynomial of degree n is the characteristic polynomial of some matrix in the qualitative class of .
Henceforth, we will denote the columns of a square matrix by .
Let and be any two square matrices of order. We denote and , the matrices formed by using the columns of .
Lemma 4.2. For any two square matrices and of order ,
Proof. With the above notations, the matrix Therefore, the characteristic polynomial of is
Hence the result.
In the above lemma, the sum of the coefficients of the terms on the right hand side of an expression in EquationEquation 3(3) (3) is 1.
Example 4.3. Let be a sign pattern matrix of order 2. Then, is spectrally arbitrary by Catral et al. (Citation2009). Let , be square matrices of order in the qualitative class of . Then, we can observe that , . Also, . Note that matrices are in , a qualitative class of . The matrices , and correspond to the vectors , and , respectively.
Theorem 4.4 (Jadhav & Deore, Citation2022). Let be a potentially nilpotent sign pattern matrix of order and let be the unit vectors along the axes. Suppose there exist at least matrices which are realization of these vectors corresponding to a sign pattern . If matrices corresponding to vectors surrounding each hyperoctant differ only in one fixed row (or column), then the sign pattern is spectrally arbitrary. Moreover, any particular non-nilpotent matrix realization can be constructed as an affine combination of matrices corresponding to a hyperoctant (i.e., the unit vectors).
We can visualize all these four vectors in the above figure, wherein the region bounded by the quadrilateral contains the origin in its interior. Moreover, any two matrices corresponding to the adjacent vertices in Figure 4.3 vary only in a column. Thus, by Theorem 4.4, a sign pattern is spectrally arbitrary.
Theorem 4.5. Let be a sign pattern matrix of order . Let such that the vectors corresponding to are . Then, there exists a curve in joining the points and , such that every point on this curve is realizable by a sign pattern matrix .
Proof. Let and be the matrices correspond to the vectors in . Then from Lemma 4.2, for . If we consider the same affine combination of the corresponding vectors, then we get a vector , which has a matrix realization in for each . Hence, for establishes the required realizable curve in joining the points and .
In Figure , the curve traced by lies in the affine combination of the vectors .
Let and be any two square matrices of order . We denote and , the matrices formed by using columns of matrices and . Then, we have the following.
Lemma 4.6. Let and be any two square matrices of order . Then
Proof. Proof follows by the multilinearity property of the determinant function, similar to proof of Lemma 4.2.
We observe that the sum of the coefficients of the terms from right hand side of EquationEquation 4(4) (4) is
Let be any square sign pattern matrix of order and . Then, the matrices constructed from and as above are also in . As per the correspondence, suppose these eight matrices and correspond to vectors and in , respectively. If we plot all these eight vectors as vertices, and two vertices are adjacent if and only if the corresponding matrices vary only in one column, then we get a graph isomorphic to the following graph in as shown in Figure .
From Lemma 2.3, as the matrices corresponding to the adjacent vertices vary only in a column, all the vectors lying on edges of the above graph are realizable by a sign pattern matrix . In view of Lemma 4.6, a vector inside the convex linear combination of vectors of the type , for some is realizable by a sign pattern matrix . Therefore, there exists a curve joining the points and such that every point on the curve is realizable by a sign pattern matrix .
Example 4.7. Consider , a square sign pattern matrix of order . Let and be two matrices in . Then, If we choose , then a vector becomes which has a matrix realization as .
Let and be square matrices of order , say and . Construct an another matrix by using columns of and . More specifically, the column of the matrix is either or for , then there are such possible matrices. For , let us denote be the matrix, whose column is the column of the matrix , for .
Theorem 4.8. Let and be two square matrices of order . Then
where and .
Proof. Proof follows by multilinearity property of the determinant function.
We would like to mention that Lemma 2.3 is a special case of Theorem 4.8. If is a square sign pattern matrix of order and the matrices , then the number of matrices obtained from and as above is . All these matrices are the members of . Moreover, these matrices correspond to vectors in . Two of these vectors can be joined by an edge if the corresponding matrices vary only in a column so that we get a graph on vertices in which degree of each vertex is at least . We can observe that a point on every edge is realizable by a sign pattern matrix , and also a vector which can be expressed as an affine combination of the type as in Theorem 4.8 for some is also realizable.
Theorem 4.9. Let be a sign pattern matrix of order . Let and let and be vectors in corresponding to matrices and , respectively. Then, there exists a curve with every point on that curve is realizable by a sign pattern matrix .
Let and be any three square matrices of order over the set of real numbers. Forming a matrix by using and , where the first column of is the first column of or or . Similarly, the second and third columns of is the second and third respective columns of or or . Then, there are such possibilities for matrix . Let us denote be the matrix whose first column is the first column of , second column is the second column of and the third column is the third column of matrix where , e.g., the matrix , etc.
Theorem 4.10. Let and be any three matrices of order . Then
Proof. Proof follows by multilinearity property of the determinant function.
Let be a sign pattern matrix of order 3 and let and be any three matrices lying in . Let be matrices as defined above. Note that . Assume that for each , the matrix corresponds to the vector in . Consider a graph in with vertices and two of the vertices are joined by an edge if and only if the corresponding matrices differ only in one column. Then, we get a graph isomorphic to a graph on vertices with a degree of each vertex is at least . Every point on this edge is realizable by a matrix in . Also, a vector which can be expressed as an affine combination of the type as in EquationEquation 5(5) (5) of Theorem 4.10 for some and is realizable. In this case, we may get a degree of freedom at most . Thus, we have the following statement true.
Theorem 4.11. Let be a sign pattern of order . Let and be any three matrices in . Assume that matrices and correspond to vectors and , respectively. Then there exists a curve or a surface in such that every point on that curve or surface is realizable by a sign pattern .
In general, if we have three matrices of order , then we have the following result.
Theorem 4.12. Let and be any three matrices of order . Then
Let be a sign pattern matrix of order , let and be any three matrices in and let matrices be constructed as above. Note that all these matrices are in . Assume that the matrix corresponds to a vector in for each . If we consider a graph with vertices as these points and connect two of these vertices by an edge if and only if the corresponding matrices vary only in a column. Then, the graph will have at least edges. Note that every point on this edge is realizable by a sign pattern . Also, a vector which can be expressed as an affine combination of the type as in EquationEquation 6(6) (6) of Theorem 4.12, for some and , is realizable. Observe that degree of freedom is at most .
Theorem 4.13. Let be a sign pattern matrix of order . Let and be any three matrices in . Assume that matrices and correspond to the vectors and respectively. Then, there exists a surface of dimension at most in such that every point on that surface is realizable by a sign pattern .
In general, we can consider matrices say of order , where . For each , a matrix is constructed by using , i.e., the column of is the column of for . So we get such possible matrices.
Theorem 4.14. With the above notations, we have
Let be any sign pattern matrix of order and be any matrices in . Then, we get at most dimensional surface in with every point on that surface is realizable by a sign pattern matrix .
Theorem 4.15. Let be a sign pattern matrix of order . Let be any matrices in where . Then,there exists a surface of dimension at most in such that every point on that surface is realizable by a sign pattern .
It should be noted that if is a sign pattern matrix of order and are any matrices lying in where , then there exists at most -dimensional surface in such that every point in that surface is realizable by a sign pattern .
Let be a sign pattern matrix of order . Let and be any two matrices in . Let be the matrix whose column is the column of the matrix if otherwise is the column of the matrix , for all .
Theorem 4.16. With the above notations, we have
where the hat notation denotes the deletion of that entry from the sequence.
Proof. The proof follows by multilinearity property of the determinant function.
Theorem 4.17. Let be a sign pattern matrix of order , and be any two matrices in and be matrices as defined above for . Then
where the hat notation denotes the deletion of that entry from the sequence and each satisfies and .
Suppose the matrix has the corresponding vector for each , then every vector in which satisfies an affine linear combination as given in EquationEquation 9(9) (9) of Theorem 4.17 will also belong to . This implies that there exists at most dimensional surface in a convex linear combination of these vectors such that every point on that surface is realizable by the matrix in .
Definition 4.18. If the surface generated by the vectors by using the affine linear combination as given in EquationEquation 9(9) (9) of Theorem 4.17 has dimension , then it is called a solid, we denote this solid by .
If we plot a graph in with vertices for , and two of the vertices are connected by an edge if and only if the corresponding matrices differ only in one column, then we get a graph on vertices with degree of each vertex is at least . For a fixed value of either or , the vertices for all form the vertices of one of the faces. Theorem 4.17 is valid for corresponding to these matrices as well.
Example 4.19. Consider a sign pattern matrix . Let and be two matrices from . Then, we get the set of eight vectors . It is easy to verify that vectors are co-linear and vectors are also co-linear as shown in the following Figure . Hence, these sets of eight vectors span the two-dimensional surface in .
The above example shows that these vectors in may span lesser than dimensional surface.
Definition 4.20. Let be a sign pattern matrix of order . We shall denote the set of all realized vectors of a sign pattern by and is defined as
Lemma 4.21. Let be a sign pattern matrix of order . If , then also lies in , for all .
Proof. As , so there exists a matrix realization say for the vector in . Therefore, the characteristic polynomial of a matrix is . But then the characteristic polynomial of would be for all . Thus, the vector has a matrix realization for all in . Hence, for all .
Using Lemma 4.21, we can give a sufficient condition for a sign pattern matrix to be spectrally arbitrary.
Theorem 4.22. Let be a potentially nilpotent sign pattern matrix of order . If every point on the unit sphere lies in , then is spectrally arbitrary.
Proof. It is enough to prove that every non-zero point in lies in . For that, let be any non-zero point in . Consider the curve for . This curve will intersect the unit sphere say at . Therefore, by the hypothesis . Thus by Lemma 4.21, .
It should be noted that if is a spectrally arbitrary sign pattern matrix, then obviously . Thus, the above theorem establishes a necessary and sufficient condition for a potentially nilpotent sign pattern matrix to be spectrally arbitrary. We have used the unit sphere in the above theorem. However, any -dimensional closed surface which encloses the origin in its interior would also work. If any such a closed surface lies in , then the unit sphere would also belong to .
Theorem 4.23. Let be a nilpotent sign pattern matrix of order , and let and be any two matrices such that they generate a solid lying in . If contains the origin in its interior, then a sign pattern is spectrally arbitrary.
Proof. Such a dimensional solid lying in the qualitative class of has its closed boundary surface of dimension with the origin lying in its interior. Then, belongs to the qualitative class of , and thus by Theorem 4.22, a sign pattern is spectrally arbitrary.
Finally, we discuss a very general case. Let be a sign pattern matrix of order . Let be any matrices belonging to the qualitative class of . Let be the matrix whose column is the column of matrix for and . Then, there are such possible matrices. Similar to Theorem 4.17.
Theorem 4.24. With the above notations, we have
Proof. Proof basically uses the multi-linearity property of the determinant function.
Assume that the matrix corresponds to the vector in . We can consider a graph with these vectors as points in , and two of these points are joined if and only if matrices corresponding to them vary only in one column. So we get a graph containing a sub-graph isomorphic to the graph having vertices and degree of each vertex is at least . Theorem 4.24 says that every point which satisfies the affine combination as given above will also belong to the qualitative class of . Thus, we get at most dimensional surface in lying in the qualitative class of for . If , then dimension of an affine surface generated by considering an affine combination given in Theorem 4.24 is at most . If it has dimension exactly , then it is a solid, and we denote this solid by .
Theorem 4.25. Let be a nilpotent sign pattern matrix of order . Let be any matrices such that they generate a solid lying in the qualitative class of . If contains the origin in its interior, then the sign pattern is spectrally arbitrary.
Proof. The proof is similar to the proof of Theorem 4.23.
5. Open question
In Section 4, we may raise an open question “If the set of vectors for spans , then the surface is a solid of dimension ”.
Acknowledgment
The authors would like to express their sincere gratitude to the learned referees for their valuable comments and suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Additional information
Funding
References
- Bergsma, H., Meulen, K. N. V., & Tuyl, A. V. (2012). Potentially nilpotent patterns and the Nilpotent-Jacobian method. Linear Algebra and Its Applications, 436(12), 4433–14. https://doi.org/10.1016/j.laa.2011.05.017
- Catral, M., Olesky, D. D., & van den Driessche, P. (2009). Allow problems concerning spectral properties of sign pattern matrices: A survey. Linear Algebra and Its Applications, 430(11–12), 3080–3094. https://doi.org/10.1016/j.laa.2009.01.031
- Cavers, M. (2021). Polynomial stability and potentially stable patterns. Linear Algebra and Its Applications, 613, 87–114. https://doi.org/10.1016/j.laa.2020.12.015
- Eschenbach, C., & Johnson, C. R. (1988). Research problems several open problems in qualitative matrix theory involving eigenvalue distribution. Linear and Multilinear Algebra, 24(1), 79–80. https://doi.org/10.1080/03081088808817900
- Grundy, D. A., Olesky, D. D., & van den Driessche, P. (2012). Constructions for potentially stable sign patterns. Linear Algebra and Its Applications, 436, 4473–4488. https://doi.org/10.1016/j.laa.2011.08.011
- Hogben, L., Hall, & Li. (2018). Sign pattern matrices, handbook of linear algebra. chapman and hall/CRC, Taylor and Francis Group. 2, 33.
- Jadhav, D. S., & Deore, R. P. (2022). A geometric construction for spectrally arbitrary sign pattern matrices and the 2n-conjecture. Czechoslovak Mathematical Journal.
- Luo, J., Huang, T. Z., Li, H., Li, Z., & Zhang, L. (2015). Tree sign patterns that allow nilpotence of index 4. Linear and Multilinear Algebra, 63-5, 1009–1025. https://doi.org/10.1080/03081087.2014.914930