-
On the modelling and prediction of high-dimensional functional time series
Authors:
**yuan Chang,
Qin Fang,
Xinghao Qiao,
Qiwei Yao
Abstract:
We propose a two-step procedure to model and predict high-dimensional functional time series, where the number of function-valued time series $p$ is large in relation to the length of time series $n$. Our first step performs an eigenanalysis of a positive definite matrix, which leads to a one-to-one linear transformation for the original high-dimensional functional time series, and the transformed…
▽ More
We propose a two-step procedure to model and predict high-dimensional functional time series, where the number of function-valued time series $p$ is large in relation to the length of time series $n$. Our first step performs an eigenanalysis of a positive definite matrix, which leads to a one-to-one linear transformation for the original high-dimensional functional time series, and the transformed curve series can be segmented into several groups such that any two subseries from any two different groups are uncorrelated both contemporaneously and serially. Consequently in our second step those groups are handled separately without the information loss on the overall linear dynamic structure. The second step is devoted to establishing a finite-dimensional dynamical structure for all the transformed functional time series within each group. Furthermore the finite-dimensional structure is represented by that of a vector time series. Modelling and forecasting for the original high-dimensional functional time series are realized via those for the vector time series in all the groups. We investigate the theoretical properties of our proposed methods, and illustrate the finite-sample performance through both extensive simulation and two real datasets.
△ Less
Submitted 2 June, 2024;
originally announced June 2024.
-
Autoregressive Networks with Dependent Edges
Authors:
**yuan Chang,
Qin Fang,
Eric D. Kolaczyk,
Peter W. MacDonald,
Qiwei Yao
Abstract:
We propose an autoregressive framework for modelling dynamic networks with dependent edges. It encompasses the models which accommodate, for example, transitivity, density-dependent and other stylized features often observed in real network data. By assuming the edges of network at each time are independent conditionally on their lagged values, the models, which exhibit a close connection with tem…
▽ More
We propose an autoregressive framework for modelling dynamic networks with dependent edges. It encompasses the models which accommodate, for example, transitivity, density-dependent and other stylized features often observed in real network data. By assuming the edges of network at each time are independent conditionally on their lagged values, the models, which exhibit a close connection with temporal ERGMs, facilitate both simulation and the maximum likelihood estimation in the straightforward manner. Due to the possible large number of parameters in the models, the initial MLEs may suffer from slow convergence rates. An improved estimator for each component parameter is proposed based on an iteration based on the projection which mitigates the impact of the other parameters (Chang et al., 2021, 2023). Based on a martingale difference structure, the asymptotic distribution of the improved estimator is derived without the stationarity assumption. The limiting distribution is not normal in general, and it reduces to normal when the underlying process satisfies some mixing conditions. Illustration with a transitivity model was carried out in both simulation and a real network data set.
△ Less
Submitted 24 April, 2024;
originally announced April 2024.
-
Approximate Core Allocations for Edge Cover Games
Authors:
Tianhang Lu,
Han Xian,
Qizhi Fang
Abstract:
We study the approximate core for edge cover games, which are cooperative games stemming from edge cover problems. In these games, each player controls a vertex on a network $G = (V, E; w)$, and the cost of a coalition $S\subseteq V$ is equivalent to the minimum weight of edge covers in the subgraph induced by $S$. We prove that the 3/4-core of edge cover games is always non-empty and can be compu…
▽ More
We study the approximate core for edge cover games, which are cooperative games stemming from edge cover problems. In these games, each player controls a vertex on a network $G = (V, E; w)$, and the cost of a coalition $S\subseteq V$ is equivalent to the minimum weight of edge covers in the subgraph induced by $S$. We prove that the 3/4-core of edge cover games is always non-empty and can be computed in polynomial time by using linear program duality approach. This ratio is the best possible, as it represents the integrality gap of the natural LP for edge cover problems. Moreover, our analysis reveals that the ratio of approximate core corresponds with the length of the shortest odd cycle of underlying graphs.
△ Less
Submitted 22 August, 2023;
originally announced August 2023.
-
Adaptive Functional Thresholding for Sparse Covariance Function Estimation in High Dimensions
Authors:
Qin Fang,
Shaojun Guo,
Xinghao Qiao
Abstract:
Covariance function estimation is a fundamental task in multivariate functional data analysis and arises in many applications. In this paper, we consider estimating sparse covariance functions for high-dimensional functional data, where the number of random functions p is comparable to, or even larger than the sample size n. Aided by the Hilbert--Schmidt norm of functions, we introduce a new class…
▽ More
Covariance function estimation is a fundamental task in multivariate functional data analysis and arises in many applications. In this paper, we consider estimating sparse covariance functions for high-dimensional functional data, where the number of random functions p is comparable to, or even larger than the sample size n. Aided by the Hilbert--Schmidt norm of functions, we introduce a new class of functional thresholding operators that combine functional versions of thresholding and shrinkage, and propose the adaptive functional thresholding estimator by incorporating the variance effects of individual entries of the sample covariance function into functional thresholding. To handle the practical scenario where curves are partially observed with errors, we also develop a nonparametric smoothing approach to obtain the smoothed adaptive functional thresholding estimator and its binned implementation to accelerate the computation. We investigate the theoretical properties of our proposals when p grows exponentially with n under both fully and partially observed functional scenarios. Finally, we demonstrate that the proposed adaptive functional thresholding estimators significantly outperform the competitors through extensive simulations and the functional connectivity analysis of two neuroimaging datasets.
△ Less
Submitted 14 July, 2022;
originally announced July 2022.
-
Alternating runs of permutations and the central factorial numbers
Authors:
Qi Fang,
Ya-Nan Feng,
Shi-Mei Ma
Abstract:
Let R(n,k) be the number of permutations of $\{1,2,\ldots,n\}$ with k alternating runs. In this paper, we establish the relationships between R(n,k) and the central factorial numbers of even indices as well as the number of signed permutations with a given number of alternating runs and the central factorial numbers of odd indices. The explicit formulas of the peak and left peak polynomials for pe…
▽ More
Let R(n,k) be the number of permutations of $\{1,2,\ldots,n\}$ with k alternating runs. In this paper, we establish the relationships between R(n,k) and the central factorial numbers of even indices as well as the number of signed permutations with a given number of alternating runs and the central factorial numbers of odd indices. The explicit formulas of the peak and left peak polynomials for permutations and the derivative polynomials of the tangent and secant functions are also established.
△ Less
Submitted 3 March, 2022; v1 submitted 28 February, 2022;
originally announced February 2022.
-
Alternating Eulerian polynomials and left peak polynomials
Authors:
Shi-Mei Ma,
Qi Fang,
Toufik Mansour,
Yeong-Nan Yeh
Abstract:
In this paper we present grammatical interpretations of the alternating Eulerian polynomials of types A and B. As applications, we derive several properties of the type B alternating Eulerian polynomials, including combinatorial expansions, recurrence relations and generating functions. We establish an interesting connection between alternating Eulerian polynomials of type B and left peak polynomi…
▽ More
In this paper we present grammatical interpretations of the alternating Eulerian polynomials of types A and B. As applications, we derive several properties of the type B alternating Eulerian polynomials, including combinatorial expansions, recurrence relations and generating functions. We establish an interesting connection between alternating Eulerian polynomials of type B and left peak polynomials of permutations in the symmetric group, which implies that the type B alternating Eulerian polynomials have gamma-vectors alternate in sign.
△ Less
Submitted 24 June, 2021; v1 submitted 19 April, 2021;
originally announced April 2021.
-
Finite Sample Theory for High-Dimensional Functional/Scalar Time Series with Applications
Authors:
Qin Fang,
Shaojun Guo,
Xinghao Qiao
Abstract:
Statistical analysis of high-dimensional functional times series arises in various applications. Under this scenario, in addition to the intrinsic infinite-dimensionality of functional data, the number of functional variables can grow with the number of serially dependent observations. In this paper, we focus on the theoretical analysis of relevant estimated cross-(auto)covariance terms between tw…
▽ More
Statistical analysis of high-dimensional functional times series arises in various applications. Under this scenario, in addition to the intrinsic infinite-dimensionality of functional data, the number of functional variables can grow with the number of serially dependent observations. In this paper, we focus on the theoretical analysis of relevant estimated cross-(auto)covariance terms between two multivariate functional time series or a mixture of multivariate functional and scalar time series beyond the Gaussianity assumption. We introduce a new perspective on dependence by proposing functional cross-spectral stability measure to characterize the effect of dependence on these estimated cross terms, which are essential in the estimates for additive functional linear regressions. With the proposed functional cross-spectral stability measure, we develop useful concentration inequalities for estimated cross-(auto)covariance matrix functions to accommodate more general sub-Gaussian functional linear processes and, furthermore, establish finite sample theory for relevant estimated terms under a commonly adopted functional principal component analysis framework. Using our derived non-asymptotic results, we investigate the convergence properties of the regularized estimates for two additive functional linear regression applications under sparsity assumptions including functional linear lagged regression and partially functional linear regression in the context of high-dimensional functional/scalar time series.
△ Less
Submitted 31 July, 2021; v1 submitted 16 April, 2020;
originally announced April 2020.
-
Polynomial control on weighted stability bounds and inversion norms of localized matrices on simple graphs
Authors:
Qiquan Fang,
Chang Eon Shin,
Qiyu Sun
Abstract:
The (un)weighted stability for some matrices is one of essential hypotheses in time-frequency analysis and applied harmonic analysis. In the first part of this paper, we show that for a localized matrix in a Beurling algebra, its weighted stabilities for different exponents and Muckenhoupt weights are equivalent to each other, and reciprocal of its optimal lower stability bound for one exponent an…
▽ More
The (un)weighted stability for some matrices is one of essential hypotheses in time-frequency analysis and applied harmonic analysis. In the first part of this paper, we show that for a localized matrix in a Beurling algebra, its weighted stabilities for different exponents and Muckenhoupt weights are equivalent to each other, and reciprocal of its optimal lower stability bound for one exponent and weight is controlled by a polynomial of reciprocal of its optimal lower stability bound for another exponent and weights. Inverse-closed Banach subalgebras of matrices with certain off-diagonal decay can be informally interpreted as localization preservation under inversion, which is of great importance in many mathematical and engineering fields. Let ${\mathcal B}(\ell^p_w)$ be the Banach algebra of bounded operators on the weighted sequence space $\ell^p_w$ on a simple graph. In the second part of this paper, we prove that Beurling algebras of localized matrices on a simple graph are inverse-closed in ${\mathcal B}(\ell^p_w)$ for all $1\le p<\infty$ and Muckenhoupt $A_p$-weights $w$, and the Beurling norm of the inversion of a matrix $A$ is bounded by a bivariate polynomial of the Beurling norm of the matrix $A$ and the operator norm of its inverse $A^{-1}$ in ${\mathcal B}(\ell^p_w)$.
△ Less
Submitted 18 September, 2019;
originally announced September 2019.
-
Spectra of Bochner-Riesz means on $L^p$
Authors:
Yang Chen,
Qiquan Fang,
Qiyu Sun
Abstract:
The Bochner-Riesz means are shown to have either the unit interval $[0,1]$ or the whole complex plane as their spectra on $L^p, 1\le p<\infty$
The Bochner-Riesz means are shown to have either the unit interval $[0,1]$ or the whole complex plane as their spectra on $L^p, 1\le p<\infty$
△ Less
Submitted 10 November, 2015;
originally announced November 2015.
-
Equivalence of robust stabilization and robust performance via feedback
Authors:
J. A. Ball,
Q. Fang,
G. J. Groenewald,
S. ter Horst
Abstract:
One approach to robust control for linear plants with structured uncertainty as well as for linear parameter-varying (LPV) plants (where the controller has on-line access to the varying plant parameters) is through linear-fractional-transformation (LFT) models. Control issues to be addressed by controller design in this formalism include robust stability and robust performance. Here robust perfo…
▽ More
One approach to robust control for linear plants with structured uncertainty as well as for linear parameter-varying (LPV) plants (where the controller has on-line access to the varying plant parameters) is through linear-fractional-transformation (LFT) models. Control issues to be addressed by controller design in this formalism include robust stability and robust performance. Here robust performance is defined as the achievement of a uniform specified $L^{2}$-gain tolerance for a disturbance-to-error map combined with robust stability. By setting the disturbance and error channels equal to zero, it is clear that any criterion for robust performance also produces a criterion for robust stability. Counter-intuitively, as a consequence of the so-called Main Loop Theorem, application of a result on robust stability to a feedback configuration with an artificial full-block uncertainty operator added in feedback connection between the error and disturbance signals produces a result on robust performance. The main result here is that this performance-to-stabilization reduction principle must be handled with care for the case of dynamic feedback compensation: casual application of this principle leads to the solution of a physically uninteresting problem, where the controller is assumed to have access to the states in the artificially-added feedback loop. Application of the principle using a known more refined dynamic-control robust stability criterion, where the user is allowed to specify controller partial-state dimensions, leads to correct robust-performance results. These latter results involve rank conditions in addition to Linear Matrix Inequality (LMI) conditions.
△ Less
Submitted 19 August, 2008;
originally announced August 2008.
-
Multivariable generalizations of the Schur class: positive kernel characterization and transfer function realization
Authors:
Joseph A. Ball,
Animikh Biswas,
Quanlei Fang,
Sanne ter Horst
Abstract:
The operator-valued Schur-class is defined to be the set of holomorphic functions $S$ map** the unit disk into the space of contraction operators between two Hilbert spaces. There are a number of alternate characterizations: the operator of multiplication by $S$ defines a contraction operator between two Hardy Hilbert spaces, $S$ satisfies a von Neumann inequality, a certain operator-valued ke…
▽ More
The operator-valued Schur-class is defined to be the set of holomorphic functions $S$ map** the unit disk into the space of contraction operators between two Hilbert spaces. There are a number of alternate characterizations: the operator of multiplication by $S$ defines a contraction operator between two Hardy Hilbert spaces, $S$ satisfies a von Neumann inequality, a certain operator-valued kernel associated with $S$ is positive-definite, and $S$ can be realized as the transfer function of a dissipative (or even conservative) discrete-time linear input/state/output linear system. Various multivariable generalizations of this class have appeared recently,one of the most encompassing being that of Muhly and Solel where the unit disk is replaced by the strict unit ball of the elements of a dual correspondence $E^σ$ associated with a $W^{*}$-correspondence $E$ over a $W^{*}$-algebra $\cA$ together with a $*$-representation $σ$ of $\cA$. The main new point which we add here is the introduction of the notion of reproducing kernel Hilbert correspondence and identification of the Muhly-Solel Hardy spaces as reproducing kernel Hilbert correspondences associated with a completely positive analogue of the classical Szegö kernel. In this way we are able to make the analogy between the Muhly-Solel Schur class and the classical Schur class more complete. We also illustrate the theory by specializing it to some well-studied special cases; in some instances there result new kinds of realization theorems.
△ Less
Submitted 2 November, 2007; v1 submitted 14 May, 2007;
originally announced May 2007.
-
Schur-class multipliers on the Arveson space: de Branges-Rovnyak reproducing kernel spaces and commutative transfer-function realizations
Authors:
Joseph A. Ball,
Vladimir Bolotnikov,
Quanlei Fang
Abstract:
An interesting and recently much studied generalization of the classical Schur class is the class of contractive operator-valued multipliers $S$ for the reproducing kernel Hilbert space ${\mathcal H}(k_{d})$ on the unit ball ${\mathbb B}^{d} \subset {\mathbb C}^{d}$, where $k_{d}$ is the positive kernel $k_{d}(λ, ζ) = 1/(1 - < λ, ζ>)$ on ${\mathbb B}^{d}$. The reproducing kernel space…
▽ More
An interesting and recently much studied generalization of the classical Schur class is the class of contractive operator-valued multipliers $S$ for the reproducing kernel Hilbert space ${\mathcal H}(k_{d})$ on the unit ball ${\mathbb B}^{d} \subset {\mathbb C}^{d}$, where $k_{d}$ is the positive kernel $k_{d}(λ, ζ) = 1/(1 - < λ, ζ>)$ on ${\mathbb B}^{d}$. The reproducing kernel space ${\mathcal H}(K_{S})$ associated with the positive kernel $K_{S}(λ, ζ) = (I - S(λ) S(ζ)^{*}) \cdot k_{d}(λ, ζ)$ is a natural multivariable generalization of the classical de Branges-Rovnyak canonical model space. A special feature appearing in the multivariable case is that the space ${\mathcal H}(K_{S})$ in general may not be invariant under the adjoints $M_{λ_{j}}^{*}$ of the multiplication operators $M_{λ_{j}} \colon f(λ) \mapsto λ_{j} f(λ)$ on ${\mathcal H}(k_{d})$.
We show that invariance of ${\mathcal H}(K_{S})$ under $M_{λ_{j}}^{*}$ for each $j = 1, ..., d$ is equivalent to the existence of a weakly coisometric realization for $S$ of the form $S(λ) = D + C (I - λ_{1}A_{1} ... - λ_{d} A_{d})^{-1}(λ_{1}B_{1} + ... + λ_{d} B_{d})$ such that the state operators $A_{1}, ..., A_{d}$ pairwise commute. We show that this special situation always occurs for the case of inner functions $S$ (where the associated multiplication operator $M_{S}$ is a partial isometry), and that inner multipliers are characterized by the existence of such a realization such that the state operators $A_{1}, >..., A_{d}$ satisfy an additional stability property.
△ Less
Submitted 20 October, 2006;
originally announced October 2006.
-
Transfer-function realization for multipliers of the Arveson space
Authors:
Joseph A. Ball,
Vladimir Bolotnikov,
Quanlei Fang
Abstract:
An interesting and recently much studied generalization of the classical Schur class is the class of contractive operator-valued multipliers for the reproducing kernel Hilbert space ${\mathcal H}(k_{d})$ on the unit ball ${\mathbb B}^{d} \subset {\mathbb C}^{d}$, where $k_{d}$ is the positive kernel $k_{d}(λ, ζ) = 1/(1 - < λ, ζ>)$ on ${\mathbb B}^{d}$. We study this space from the point of view…
▽ More
An interesting and recently much studied generalization of the classical Schur class is the class of contractive operator-valued multipliers for the reproducing kernel Hilbert space ${\mathcal H}(k_{d})$ on the unit ball ${\mathbb B}^{d} \subset {\mathbb C}^{d}$, where $k_{d}$ is the positive kernel $k_{d}(λ, ζ) = 1/(1 - < λ, ζ>)$ on ${\mathbb B}^{d}$. We study this space from the point of view of realization theory and functional models of de Branges-Rovnyak type. We highlight features which depart from the classical univariate case: coisometric realizations have only partial uniqueness properties, the nonuniqueness can be described explicitly, and this description assumes a particularly concrete form in the functional-model context.
△ Less
Submitted 20 October, 2006;
originally announced October 2006.
-
Schur-class multipliers on the Fock space: de Branges-Rovnyak reproducing kernel spaces and transfer-function realizations
Authors:
Joseph A. Ball,
Vladimir Bolotnikov,
Quanlei Fang
Abstract:
We introduce and study a Fock-space noncommutative analogue of reproducing kernel Hilbert spaces of de Branges-Rovnyak type. Results include: use of the de Branges-Rovnyak space ${\mathcal H}(K_{S})$ as the state space for the unique (up to unitary equivalence) observable, coisometric transfer-function realization of the Schur-class multiplier $S$, realization-theoretic characterization of inner…
▽ More
We introduce and study a Fock-space noncommutative analogue of reproducing kernel Hilbert spaces of de Branges-Rovnyak type. Results include: use of the de Branges-Rovnyak space ${\mathcal H}(K_{S})$ as the state space for the unique (up to unitary equivalence) observable, coisometric transfer-function realization of the Schur-class multiplier $S$, realization-theoretic characterization of inner Schur-class multipliers, and a calculus for obtaining a realization for an inner multiplier with prescribed left zero-structure. In contrast with the parallel theory for the Arveson space on the unit ball ${\mathbb B}^{d} \subset {\mathbb C}^{d}$ (which can be viewed as the symmetrized version of the Fock space used here), the results here are much more in line with the classical univariate case, with the extra ingredient of the existence of all results having both a ``left'' and a ``right'' version.
△ Less
Submitted 20 October, 2006;
originally announced October 2006.
-
Multivariable backward-shift-invariant subspaces and observability operators
Authors:
Joseph A. Ball,
Vladimir Bolotnikov,
Quanlei Fang
Abstract:
It is well known that subspaces of the Hardy space over the unit disk which are invariant under the backward shift occur as the image of an observability operator associated with a discrete-time linear system with stable state-dynamics, as well as the functional-model space for a Hilbert space contraction operator. We discuss two multivariable extensions of this structure, where the classical Ha…
▽ More
It is well known that subspaces of the Hardy space over the unit disk which are invariant under the backward shift occur as the image of an observability operator associated with a discrete-time linear system with stable state-dynamics, as well as the functional-model space for a Hilbert space contraction operator. We discuss two multivariable extensions of this structure, where the classical Hardy space is replaced by (1) the Fock space of formal power series in a collection of $d$ noncommuting indeterminates with norm-square-summable vector coefficients, and (2) the reproducing kernel Hilbert space (often now called the Arveson space) over the unit ball in ${\mathbb C}^{d}$ with reproducing kernel $k(λ, ζ) = 1/(1 - < λ, ζ>)$ ($\lam, ζ\in {\mathbb C}^{d}$ with $\| λ\|, \| ζ\| < 1$). In the first case, the associated linear system is of noncommutative Fornasini-Marchesini type with evolution along a free semigroup with $d$ generators, while in the second case the linear system is a standard (commutative) Fornasini-Marchesini-type system with evolution along the integer lattice ${\mathbb Z}^{d}$. An abelianization map (or symmetrization of the Fock space) links the first case with the second. The second case has special features depending on whether the operator-tuple defining the state dynamics is commutative or not. The paper focuses on multidimensional state-output linear systems and the associated observability operators.
△ Less
Submitted 20 October, 2006;
originally announced October 2006.
