Asymptotic Linearity Research Articles

Diffusion minimum-Wilcoxon-norm over distributed adaptive networks: Formulation and performance analysis

This paper deals with the development of robust diffusion strategy for wireless sensor networks using minimum-Wilcoxon-norm. The Wilcoxon norm based robust estimation now-a-days has drawn the attention of the signal processing community for its scale equivariant property and simplicity. Exhaustive mathematical analysis has been presented to obtain the proposed diffusion minimum-Wilcoxon-norm (dMWN) algorithm. Steady state performance analysis of the algorithm has also been carried out to show the stability of the proposed method. Asymptotic linearity of rank test [1] is used for the convergence analysis of the proposed method. Extensive simulations have been given to demonstrate the efficacy of the proposed algorithm.

System Degeneracy and the Output Feedback Problem: Parametrisation of the Family of Degenerate Compensators

The paper provides a new characterisation of constant and dynamic degenerate compensators for proper multivariable systems. The motivation stems from the very important property that degenerate feedback gains may be used for the linearisation of the pole assignment map and enable frequency assignment. The objective is the characterisation and parametrisation of all feedback gains that may allow the asymptotic linearisation of the pole placement map. Such a parametrisation introduces new degrees of freedom for the linearisation of the related frequency assignment map and plays an important role to the solvability of the output feedback pole assignment problem. The paper reviews the Global Asymptotic Linearisation method associated with the core versions of determinantal pole assignment problems and defines the conditions which characterises degenerate solutions of different types. Using the theory of ordered minimal bases, we provide a parametrisation of special families of degenerate compensators according to their degree. Finally, the special properties of degenerate solutions that allow frequency assignment are considered.

Selection of Decentralised Schemes and Parametrisation of the Decentralised Degenerate Compensators

The design of decentralised control schemes has two major aspects. The selection of the decentralised structure and then the design of the decentralised controller that has a given structure and addresses certain design requirements. This paper deals with the parametrisation and selection of the decentralized structure such that problems such as the decentralised pole assignment may have solutions. We use the approach of global linearisation for the asymptotic linearisation of the pole assignment map around a degenerate compensator. Thus, we examine in depth the case of degenerate compensators and investigate the conditions under which certain degenerate structures exist. This leads to a parametrisation of decentralised structures based on the structural properties of the system.

Proof of the Asymptotic Linearity Theorem Extension Conjecture (ALTEC)

The Asymptotic Linearity Theorem, which proves the Fukui conjecture in a broader context, plays a significant role in the repeat space theory, which is the central unifying theory in the first and the second generation Fukui Project. Proving the Asymptotic Linearity Theorem Extension Conjecture (ALTEC) is a fundamental problem in the repeat space theory. In the present paper, we give a proof of the ALTEC using inwardly repeating fractal structures of some functions used in Matrix Art from the Niagara Project, which is part of the second generation Fukui Project. In recent years, the Niagara Project has been extended to the broader project of Science-Art Multi-angle Network, and the second generation Fukui Project is also called the New Frontier Project.

Second-Order Inference for the Mean of a Variable Missing at Random.

We present a second-order estimator of the mean of a variable subject to missingness, under the missing at random assumption. The estimator improves upon existing methods by using an approximate second-order expansion of the parameter functional, in addition to the first-order expansion employed by standard doubly robust methods. This results in weaker assumptions about the convergence rates necessary to establish consistency, local efficiency, and asymptotic linearity. The general estimation strategy is developed under the targeted minimum loss-based estimation (TMLE) framework. We present a simulation comparing the sensitivity of the first and second-order estimators to the convergence rate of the initial estimators of the outcome regression and missingness score. In our simulation, the second-order TMLE always had a coverage probability equal or closer to the nominal value 0.95, compared to its first-order counterpart. In the best-case scenario, the proposed second-order TMLE had a coverage probability of 0.86 when the first-order TMLE had a coverage probability of zero. We also present a novel first-order estimator inspired by a second-order expansion of the parameter functional. This estimator only requires one-dimensional smoothing, whereas implementation of the second-order TMLE generally requires kernel smoothing on the covariate space. The first-order estimator proposed is expected to have improved finite sample performance compared to existing first-order estimators. In the best-case scenario of our simulation study, the novel first-order TMLE improved the coverage probability from 0 to 0.90. We provide an illustration of our methods using a publicly available dataset to determine the effect of an anticoagulant on health outcomes of patients undergoing percutaneous coronary intervention. We provide R code implementing the proposed estimator.

On Zero-Delay Source-Channel Coding

In this paper, we study the zero-delay source-channel coding problem, and specifically the problem of obtaining the vector transformations that optimally map between the m-dimensional source space and the k-dimensional channel space, under a given transmission power constraint and for the mean square error distortion. We first study the functional properties of this problem and show that the objective is concave in the source and noise densities and convex in the density of the input to the channel. We then derive the necessary conditions for optimality of the encoder and decoder mappings. A well known result in information theory pertains to the linearity of optimal encoding and decoding mappings in the scalar Gaussian source and channel setting, at all channel signal-to-noise ratios (CSNRs). In this paper, we study this result more generally, beyond the Gaussian source and channel, and derive the necessary and sufficient condition for linearity of optimal mappings, given a noise (or source) distribution, and a specified power constraint. We also prove that the Gaussian source-channel pair is unique in the sense that it is the only source-channel pair for which the optimal mappings are linear at more than one CSNR values. Moreover, we show the asymptotic linearity of optimal mappings for low CSNR if the channel is Gaussian regardless of the source and, at the other extreme, for high CSNR if the source is Gaussian, regardless of the channel. Our numerical results show strict improvement over prior methods. The numerical approach is extended to the scenario of source-channel coding with decoder side information. The resulting encoding mappings are shown to be continuous relatives of, and in fact subsume as special case, the Wyner-Ziv mappings encountered in digital distributed source coding systems.

Targeted Learning of the Mean Outcome under an Optimal Dynamic Treatment Rule.

We consider estimation of and inference for the mean outcome under the optimal dynamic two time-point treatment rule defined as the rule that maximizes the mean outcome under the dynamic treatment, where the candidate rules are restricted to depend only on a user-supplied subset of the baseline and intermediate covariates. This estimation problem is addressed in a statistical model for the data distribution that is nonparametric beyond possible knowledge about the treatment and censoring mechanism. This contrasts from the current literature that relies on parametric assumptions. We establish that the mean of the counterfactual outcome under the optimal dynamic treatment is a pathwise differentiable parameter under conditions, and develop a targeted minimum loss-based estimator (TMLE) of this target parameter. We establish asymptotic linearity and statistical inference for this estimator under specified conditions. In a sequentially randomized trial the statistical inference relies upon a second-order difference between the estimator of the optimal dynamic treatment and the optimal dynamic treatment to be asymptotically negligible, which may be a problematic condition when the rule is based on multivariate time-dependent covariates. To avoid this condition, we also develop TMLEs and statistical inference for data adaptive target parameters that are defined in terms of the mean outcome under the estimate of the optimal dynamic treatment. In particular, we develop a novel cross-validated TMLE approach that provides asymptotic inference under minimal conditions, avoiding the need for any empirical process conditions. We offer simulation results to support our theoretical findings.

BOOTSTRAPPING DENSITY-WEIGHTED AVERAGE DERIVATIVES

We investigate the properties of several bootstrap-based inference procedures for semiparametric density-weighted average derivatives. The key innovation in this paper is to employ an alternative asymptotic framework to assess the properties of these inference procedures. This theoretical approach is conceptually distinct from the traditional approach (based on asymptotic linearity of the estimator and Edgeworth expansions), and leads to different theoretical prescriptions for bootstrap-based semiparametric inference. First, we show that the conventional bootstrap-based approximations to the distribution of the estimator and its classical studentized version are both invalid in general. This result shows a fundamental lack of “robustness” of the associated, classical bootstrap-based inference procedures with respect to the bandwidth choice. Second, we present a new bootstrap-based inference procedure for density-weighted average derivatives that is more “robust” to perturbations of the bandwidth choice, and hence exhibits demonstrable superior theoretical statistical properties over the traditional bootstrap-based inference procedures. Finally, we also examine the validity and invalidity of related bootstrap-based inference procedures and discuss additional results that may be of independent interest. Some simulation evidence is also presented.

Reduced Sensitivity Solutions to Global Linearisation of the Pole Assignment Map

The problem of pole assignment, by static output feedback controllers has been tackled as far as solvability conditions and the computation of solutions when they exist by a powerful method referred to as global linearisation. This is based on asymptotic linearisation (around a degenerate point) of the pole placement map. The essence of the present approach is to reduce the multilinear nature of the problem to the solution of a linear set of equations. The solution is given in closed form in terms of a one-parameter family of static feedback compensators, for which the closed-loop poles approach the required ones as ɛ → 0. The use of degenerate compensators makes the method numerically sensitive. This paper develops further the global linearisation framework by developing numerical techniques which make the method less sensitive to the use of degenerate solutions as the basis of the methodology. The proposed new computational framework for finding output feedback controllers improves considerably the sensitivity properties by using a predictor-corrector numerical method based on homotopy continuation. The modified method guarantees the maximum distance from the degenerate point. The current numerical method developed for the constant output feedback extends also to the case of dynamic output feedback.

Targeted Estimation of Nuisance Parameters to Obtain Valid Statistical Inference

In order to obtain concrete results, we focus on estimation of the treatment specific mean, controlling for all measured baseline covariates, based on observing independent and identically distributed copies of a random variable consisting of baseline covariates, a subsequently assigned binary treatment, and a final outcome. The statistical model only assumes possible restrictions on the conditional distribution of treatment, given the covariates, the so-called propensity score. Estimators of the treatment specific mean involve estimation of the propensity score and/or estimation of the conditional mean of the outcome, given the treatment and covariates. In order to make these estimators asymptotically unbiased at any data distribution in the statistical model, it is essential to use data-adaptive estimators of these nuisance parameters such as ensemble learning, and specifically super-learning. Because such estimators involve optimal trade-off of bias and variance w.r.t. the infinite dimensional nuisance parameter itself, they result in a sub-optimal bias/variance trade-off for the resulting real-valued estimator of the estimand. We demonstrate that additional targeting of the estimators of these nuisance parameters guarantees that this bias for the estimand is second order and thereby allows us to prove theorems that establish asymptotic linearity of the estimator of the treatment specific mean under regularity conditions. These insights result in novel targeted minimum loss-based estimators (TMLEs) that use ensemble learning with additional targeted bias reduction to construct estimators of the nuisance parameters. In particular, we construct collaborative TMLEs (C-TMLEs) with known influence curve allowing for statistical inference, even though these C-TMLEs involve variable selection for the propensity score based on a criterion that measures how effective the resulting fit of the propensity score is in removing bias for the estimand. As a particular special case, we also demonstrate the required targeting of the propensity score for the inverse probability of treatment weighted estimator using super-learning to fit the propensity score.

Generalized Jackknife Estimators of Weighted Average Derivatives

With the aim of improving the quality of asymptotic distributional approximations for nonlinear functionals of nonparametric estimators, this article revisits the large-sample properties of an important member of that class, namely a kernel-based weighted average derivative estimator. Asymptotic linearity of the estimator is established under weak conditions. Indeed, we show that the bandwidth conditions employed are necessary in some cases. A bias-corrected version of the estimator is proposed and shown to be asymptotically linear under yet weaker bandwidth conditions. Implementational details of the estimators are discussed, including bandwidth selection procedures. Consistency of an analog estimator of the asymptotic variance is also established. Numerical results from a simulation study and an empirical illustration are reported. To establish the results, a novel result on uniform convergence rates for kernel estimators is obtained. The online supplemental material to this article includes details on the theoretical proofs and other analytic derivations, and further results from the simulation study.

Residual empirical processes and their application to GM-testing for the autoregression order

In the paper a new nonparametric generalized M-test for hypotheses about the order of linear autoregression AR(p) is constructed. We also establish robustness of this test in the model of data contamination by independent additive outliers with intensity O(n−1/2). Robustness is formulated in terms of limiting power equicontinuity. Test statistics are constructed with the help of residual empirical processes. We establish the asymptotic uniform linearity of these processes in the defined contamination model.

On Quadratic Expansions of Log-Likelihoods and a General Asymptotic Linearity Result

Irrespective of the statistical model under study, the derivation of limits, in the Le Cam sense, of sequences of local experiments (see [7]-[10]) often follows along very similar lines, essentially involving differentiability in quadratic mean of square roots of (conditional) densities. This chapter establishes two abstract and very general results providing sufficient and nearly necessary conditions for (i) the existence of a quadratic expansion, and (ii) the asymptotic linearity of local log-likelihood ratios (asymptotic linearity is needed whenever unspecified model parameters are to be replaced, in some statistic of interest, with some preliminary estimator). Such results have been established, for locally asymptotically normal (LAN) models involving independent and identically distributed observations, by, e.g., [1], [11] and [12]. Similar results are provided here for models exhibiting serial dependencies which, so far, have been treated on a case-by-case basis (see [4] and [5] for typical examples) and, in general, under stronger regularity assumptions. Unlike their i.i.d. counterparts, our results moreover extend beyond the context of LAN experiments, so that non-stationary unit-root time series and cointegration models, for instance, also can be handled (see, e.g., [6]).

Asymptotic linearity of a linear rank statistic in the case of symmetric nonidentically distributed variables

Let Y 1, …, Y n be independent but not identically distributed random variables with densities f 1, …, f n symmetric around zero. Suppose c 1, n , …, c n, n are given constants such that ∑ i c i, n =0 and . Denote the rank of Y i −Δ c i, n for any Δ∈ℝ by R(Y i −Δ c i, n ) and let a n (i) be a score defined via a score function ϕ. We study the linear rank statistic and prove that S n (Δ) is asymptotically uniformly linear in the parameter Δ in any interval [−C, C], C>0.

Repeat space theory applied to carbon nanotubes and related molecular networks. III

The present article is the first part of a series devoted to extending the Repeat Space Theory (RST) to apply to carbon nanotubes and related molecular networks. Four key problems are formulated whose affirmative solutions imply the formation of the initial investigative bridge between the research field of nanotubes and that of the additivity and other network problems studied and solved by using the RST. All of these four problems are solved affirmatively by using tools from the RST. The Piecewise Monotone Lemmas (PMLs) are cornerstones of the proof of the Fukui conjecture concerning the additivity problems of hydrocarbons. The solution of the fourth problem gives a generalized analytical formula of the pi-electron energy band curves of nanotube (a, b), with two new complex parameters c and d. These two parameters bring forth a broad class of analytic curves to which the PMLs and associated theoretical devices apply. Based on the above affirmative solutions of the problems, a central theorem in the RST, called the asymptotic linearity theorem (ALT) has been applied to nanotubes and monocyclic polyenes. Analytical formulae derived in this application of the ALT illuminate in a new global context (i) the conductivity of nanotubes and (ii) the aromaticity of monocyclic polyenes; moreover an analytical formula obtained by using the ALT provides a fresh insight into H¨ uckel’s (4n +2) rule. The present article forms a foundation of the forthcoming articles in this series.

On Conditions for Linearity of Optimal Estimation

When is optimal estimation linear? It is well known that when a Gaussian source is contaminated with Gaussian noise, a linear estimator minimizes the mean square estimation error. This paper analyzes, more generally, the conditions for linearity of optimal estimators. Given a noise (or source) distribution, and a specified signal-to-noise ratio (SNR), we derive conditions for existence and uniqueness of a source (or noise) distribution for which the Lp optimal estimator is linear. We then show that if the noise and source variances are equal, then the matching source must be distributed identically to the noise. Moreover, we prove that the Gaussian source-channel pair is unique in the sense that it is the only source-channel pair for which the mean square error (MSE) optimal estimator is linear at more than one SNR values. Furthermore, we show the asymptotic linearity of MSE optimal estimators for low SNR if the channel is Gaussian regardless of the source and, vice versa, for high SNR if the source is Gaussian regardless of the channel. The extension to the vector case is also considered where besides the conditions inherited from the scalar case, additional constraints must be satisfied to ensure linearity of the optimal estimator.

Pseudo-Gaussian and rank-based optimal tests for random individual effects in large [formula omitted] small [formula omitted] panels

We consider the problem of detecting unobserved heterogeneity, that is, the problem of testing the absence of random individual effects in an n×T panel. We establish a local asymptotic normality property–with respect to intercept, regression coefficient, the scale parameter σ of the error, and the scale parameter σu of individual effects (which is the parameter of interest)–for given (scaled) density f1 of the error terms, when n tends to infinity and T is fixed. This result allows, via the Hájek representation theorem, for developing asymptotically optimal rank-based tests for the null hypothesis σu=0 (absence of individual effects). These tests are locally asymptotically optimal at correctly specified innovation densities f1, but remain valid irrespective of the actual underlying density. The limiting distribution of our test statistics is obtained both under the null and under sequences of contiguous alternatives. A local asymptotic linearity property is established in order to control for the effect of substituting estimators for nuisance parameters. The asymptotic relative efficiencies of the proposed procedures with respect to the corresponding pseudo-Gaussian parametric tests are derived. In particular, the van der Waerden version of our rank-based tests uniformly dominates, from the point of view of Pitman efficiency, the classical Honda test. Small-sample performances are investigated via a Monte-Carlo study, and confirm theoretical findings.

Multidimensional Magic Mountains and Matrix Art for the generalized repeat space theory

The Asymptotic Linearity Theorem (ALT), which proves the Fukui conjecture in a broader context, plays a significant role in the repeat space theory (RST), which is the central unifying theory in the First and the Second Generation Fukui Project. Proving the Asymptotic Linearity Theorem Extension Conjecture (ALTEC) is a fundamental problem in the repeat space theory. The present paper constructs a class of functions MagicMtθ, which serves as a powerful tool for proving the Asymptotic Linearity Theorem Extension Conjecture and related propositions. The d-dimensional generalization μd,n,θ of MagicMtθ, which is given in the present paper and is called a ‘d-dimensional Magic Mountain’, provides inwardly repeating fractals in multidimensional spaces useful for interdisciplinary research that uses the generalized repeat space theory.

Perturbed solution of free non-linear vibrations of composite beams

The aim of this study is to characterize the free non-linear vibration behavior of composite beams by using polynomial finite element method with shape functions based on Legendre polynomials or sinusoidal functions. The beams are subjected to different boundary conditions. Non-linear frequencies and modes shapes are obtained using asymptotic linearization method applied on the fundamental vibrating mode. The different composite beams are subjected to different layer dispositions. The nature of the interface between layers is of no object in this study. Navier–Bernoulli hypothesis on the transverse sections is adopted. The study concerns the influence of geometrical non-linearities of the free dynamical behavior. The analysis is performed thus by considering the influence of the vibration amplitude on the modal frequencies of the structure. The solutions are obtained by numerical simulation and by semi-analytical method using asymptotic linearization applied in the real temporal modal space. The numerical values are compared to the different ones obtained in literature.

Rank-Based Tests of the Cointegrating Rank in Semiparametric Error Correction Models

This paper introduces rank-based tests for the cointegrating rank in an Error CorrectionModel with i.i.d. elliptical innovations. The tests are asymptotically distribution-free,and their validity does not depend on the actual distribution of the innovations. Thisresult holds despite the fact that, depending on the alternatives considered, the model exhibitsa non-standard Locally Asymptotically Brownian Functional (LABF) and LocallyAsymptotically Mixed Normal (LAMN) local structure—a structure which we completelycharacterize. Our tests, which have the general form of Lagrange multiplier tests, dependon a reference density that can freely be chosen, and thus is not restricted to be Gaussianas in traditional quasi-likelihood procedures. Moreover, appropriate choices of the referencedensity are achieving the semiparametric efficiency bounds. Simulations show thatour asymptotic analysis provides an accurate approximation to finite-sample behavior.Our results are based on an extension, of independent interest, of two abstract resultson the convergence of statistical experiments and the asymptotic linearity of statistics tothe context of, possibly non-stationary, time series (This abstract was borrowed from another version of this item.)