The Graphical Horseshoe Estimator for Inverse Covariance Matrices

We develop a new estimator of the inverse covariance matrix for high-dimensional multivariate normal data using the horseshoe prior. The proposed graphical horseshoe estimator has attractive properties compared to other popular estimators, such as the graphical lasso and the graphical smoothly clipped absolute deviation. The most prominent benefit is that when the true inverse covariance matrix is sparse, the graphical horseshoe provides estimates with small information divergence from the sampling model. The posterior mean under the graphical horseshoe prior can also be almost unbiased under certain conditions. In addition to these theoretical results, we also provide a full Gibbs sampler for implementing our estimator. MATLAB code is available for download from github at http://github.com/liyf1988/GHS. The graphical horseshoe estimator compares favorably to existing techniques in simulations and in a human gene network data analysis. Supplementary materials for this article are available online.     

Bayesian Dimensionality Reduction With PCA Using Penalized Semi-Integrated Likelihood

We discuss the problem of estimating the number of principal components in principal components analysis (PCA). Despite the importance of the problem and the multitude of solutions proposed in literature, it comes as a surprise that there does not exist a coherent asymptotic framework, which would justify different approaches depending on the actual size of the dataset. In this article, we address this issue by presenting an approximate Bayesian approach based on Laplace approximation and introducing a general method of developing criteria for model selection, called PEnalized SEmi-integrated Likelihood (PESEL). Our general framework encompasses a variety of existing approaches based on probabilistic models, like the Bayesian Information Criterion for Probabilistic PCA (PPCA), and enables the construction of new criteria, depending on the size of the dataset at hand and additional prior information. Specifically, we apply PESEL to derive two new criteria for datasets where the number of variables substantially exceeds the number of observations, which is out of the scope of currently existing approaches. We also report results of extensive simulation studies and real data analysis, which illustrate the desirable properties of our proposed criteria as compared to state-of-the-art methods and very recent proposals. Specifically, these simulations show that PESEL-based criteria can be quite robust against deviations from the assumptions of a probabilistic model. Selected PESEL-based criteria for the estimation of the number of principal components are implemented in the R package pesel, which is available on github (https://github.com/psobczyk/pesel). Supplementary material for this article, with additional simulation results, is available online. The code to reproduce all simulations is available at https://github.com/psobczyk/pesel_simulations.     

Tensor-on-Tensor Regression

I propose a framework for the linear prediction of a multiway array (i.e., a tensor) from another multiway array of arbitrary dimension, using the contracted tensor product. This framework generalizes several existing approaches, including methods to predict a scalar outcome from a tensor, a matrix from a matrix, or a tensor from a scalar. I describe an approach that exploits the multiway structure of both the predictors and the outcomes by restricting the coefficients to have reduced PARAFAC/CANDECOMP rank. I propose a general and efficient algorithm for penalized least-squares estimation, which allows for a ridge (L₂) penalty on the coefficients. The objective is shown to give the mode of a Bayesian posterior, which motivates a Gibbs sampling algorithm for inference. I illustrate the approach with an application to facial image data. An R package is available at https://github.com/lockEF/MultiwayRegression.     

Bayesian Synthetic Likelihood

Having the ability to work with complex models can be highly beneficial. However, complex models often have intractable likelihoods, so methods that involve evaluation of the likelihood function are infeasible. In these situations, the benefits of working with likelihood-free methods become apparent. Likelihood-free methods, such as parametric Bayesian indirect likelihood that uses the likelihood of an alternative parametric auxiliary model, have been explored throughout the literature as a viable alternative when the model of interest is complex. One of these methods is called the synthetic likelihood (SL), which uses a multivariate normal approximation of the distribution of a set of summary statistics. This article explores the accuracy and computational efficiency of the Bayesian version of the synthetic likelihood (BSL) approach in comparison to a competitor known as approximate Bayesian computation (ABC) and its sensitivity to its tuning parameters and assumptions. We relate BSL to pseudo-marginal methods and propose to use an alternative SL that uses an unbiased estimator of the SL, when the summary statistics have a multivariate normal distribution. Several applications of varying complexity are considered to illustrate the findings of this article. Supplemental materials are available online. Computer code for implementing the methods on all examples is available at https://github.com/cdrovandi/Bayesian-Synthetic-Likelihood.     

An Algorithm for Generating Individualized Treatment Decision Trees and Random Forests

With new treatments and novel technology available, precision medicine has become a key topic in the new era of healthcare. Traditional statistical methods for precision medicine focus on subgroup discovery through identifying interactions between a few markers and treatment regimes. However, given the large scale and high dimensionality of modern datasets, it is difficult to detect the interactions between treatment and high-dimensional covariates. Recently, novel approaches have emerged that seek to directly estimate individualized treatment rules (ITR) via maximizing the expected clinical reward by using, for example, support vector machines (SVM) or decision trees. The latter enjoys great popularity in clinical practice due to its interpretability. In this article, we propose a new reward function and a novel decision tree algorithm to directly maximize rewards. We further improve a single tree decision rule by an ensemble decision tree algorithm, ITR random forests. Our final decision rule is an average over single decision trees and it is a soft probability rather than a hard choice. Depending on how strong the treatment recommendation is, physicians can make decisions based on our model along with their own judgment and experience. Performance of ITR forest and tree methods is assessed through simulations along with applications to a randomized controlled trial (RCT) of 1385 patients with diabetes and an EMR cohort of 5177 patients with diabetes. ITR forest and tree methods are implemented using statistical software R (https://github.com/kdoub5ha/ITR.Forest). Supplementary materials for this article are available online.     

Asynchronously parallel optimization solver for finding multiple minima

We propose and analyze an asynchronously parallel optimization algorithm for finding multiple, high-quality minima of nonlinear optimization problems. Our multistart algorithm considers all previously evaluated points when determining where to start or continue a local optimization run. Theoretical results show that when there are finitely many minima, the algorithm almost surely starts a finite number of local optimization runs and identifies every minimum. The algorithm is applicable to general optimization settings, but our numerical results focus on the case when derivatives are unavailable. In numerical tests, a Python implementation of the algorithm is shown to yield good approximations of many minima (including a global minimum), and this ability is shown to scale well with additional resources. Our implementation’s time to solution is shown also to scale well even when the time to perform the function evaluation is highly variable. An implementation of the algorithm is available in the libEnsemble library at https://github.com/Libensemble/libensemble.     

Power boundedness in the Fourier and Fourier–Stieltjes algebras on homogeneous spaces

We study power boundedness in the Fourier and Fourier–Stieltjes algebras, Open image in new window and Open image in new window of a homogeneous space Open image in new window The main results characterizes when all elements with spectral radius at most one, in any of these algebras, are power bounded.     

Eigenfunctions for Partially Rectangular Billiards

In this note, we further develop the methods of Burq and Zworski (2005 Burq , N. , Zworski , M. ( 2005 ). Bouncing ball modes and quantum chaos . SIAM Review 47 ( 5 ): 43 – 49 [CROSSREF] [CSA] [Crossref] , [Google Scholar]) to study eigenfunctions for billiards which have rectangular components: these include the Bunimovich billiard, the Sinai billiard, and the recently popular pseudointegrable billiards (Bogomolny et al., 1999 Bogomolny , E. , Gerland , U. , Schmit , C. ( 1999 ). Models of intermediate spectral statistics . Phys. Rev. E 59 : 1315 – 1318 [CROSSREF] [CSA] [Crossref], [Web of Science ®] , [Google Scholar]). The results are an application of a “black-box” point of view as presented in Burq and Zworski (2004 Burq , N. , Zworski , M. ( 2004 ). Geometric control in the presence of a black box . JAMS 17 : 443 – 471 [CROSSREF] [CSA] [Web of Science ®] , [Google Scholar]).     

Finding small somatic structural variants in exome sequencing data: a machine learning approach

Genetic variation forms the basis for diversity but can as well be harmful and cause diseases, such as tumors. Structural variants (SV) are an example of complex genetic variations that comprise of many nucleotides ranging up to several megabases. Based on recent developments in sequencing technology it has become feasable to elucidate the genetic state of a person’s genes (i.e. the exome) or even the complete genome. Here, a machine learning approach is presented to find small disease-related SVs with the help of sequencing data. The method uses differences in characteristics of mapping patterns between tumor and normal samples at a genomic locus. This way, the method aims to be directly applicable for exome sequencing data to improve detection of SVs since specific SV detection methods are currently lacking. The method has been evaluated based on a simulation study as well as with exome data of patients with acute myeloid leukemia. An implementation of the algorithm is available at https://github.com/lenz99-/svmod.     

Self-Dual Hopf Quivers

ABSTRACT

We study self-dual coradically graded pointed Hopf algebras with a help of the dual Gabriel theorem for pointed Hopf algebras (van Oystaeyen and Zhang, 2004 van Oystaeyen , F. , Zhang , P. ( 2004 ). Quiver Hopf algebras . J. Algebra 280 ( 2 ): 577 – 589 . [CSA] [CROSSREF]  [Google Scholar]). The co-Gabriel Quivers of such Hopf algebras are said to be self-dual. An explicit classification of self-dual Hopf quivers is obtained. We also prove that finite dimensional pointed Hopf algebras with self-dual graded versions are generated by group-like and skew-primitive elements as associative algebras. This partially justifies a conjecture of Andruskiewitsch and Schneider (2000 Andruskiewitsch , N. , Schneider , H.-J . ( 2000 ). Finite quantum groups and Cartan matrices . Adv. Math. 154 : 1 – 45 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]) and may help to classify finite dimensional self-dual coradically graded pointed Hopf algebras.     

Propagation of Singularities for Schrödinger Equations on the Euclidean Space with a Scattering Metric

Given a scattering metric on the Euclidean space. We consider the Schrödinger equation corresponding to the metric, and study the propagation of singularities for the solution in terms of the “homogeneous wavefront set”. We also prove that the notion of the homogeneous wavefront set is essentially equivalent to that of the quadratic scattering wavefront set introduced by Wunsch (1999 Wunsch , J. ( 1999 ). Propagation of singularities and growth for Schrödinger operators . Duke Math. J. 98 : 137 – 186 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]). One of the main results in Wunsch (1999 Wunsch , J. ( 1999 ). Propagation of singularities and growth for Schrödinger operators . Duke Math. J. 98 : 137 – 186 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]) follows on the Euclidean space with a weaker, almost optimal condition on the potential.     

Global Stability and Bounds for Coarsening Rates Within the Mean-Field Theory for Domain Coarsening

We investigate the long-time behavior of solutions to the classical mean-field model for coarsening by Lifshitz–Slyozov and Wagner (LSW). In the original work (Lifshitz and Slyozov, 1961 Lifshitz , I. M. , Slyozov , V. V. ( 1961 ). The kinetics of precipitation from supersaturated solid solutions . J. Phys. Chem. Solids 19 : 35 – 50 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]; Wagner 1961 Wagner , C. ( 1961 ). Theorie der Alterung von Niederschlägen durch Umlösen . Z. Elektrochemie 65 : 581 – 594 . [CSA]  [Google Scholar]) convergence of solutions to a uniquely determined self-similar solution was predicted. However, it is by now well known (Giron et al., 1998 Giron , B. , Meerson , B. , Sasorov , V. P. ( 1998 ). Weak selection and stability of localized distributions in Ostwald ripening . Phys. Rev. E 58 : 4213 – 4216 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]; Niethammer and Pego 1999 Niethammer , B. , Pego , R. L. ( 1999 ). Non-self-similar behavior in the LSW theory of Ostwald ripening . J. Stat. Phys. 95 ( 5/6 ): 867 – 902 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar] 2001 Niethammer , B. , Pego , R. L. ( 2001 ). The LSW model for domain coarsening: Asymptotic behavior for total conserved mass . J. Stat. Phys. 104 ( 5/6 ): 1113 – 1144 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]) that the long-time behavior of solutions depends sensitively on the initial data. In Niethammer and Pego (1999 Niethammer , B. , Pego , R. L. ( 1999 ). Non-self-similar behavior in the LSW theory of Ostwald ripening . J. Stat. Phys. 95 ( 5/6 ): 867 – 902 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar] 2001 Niethammer , B. , Pego , R. L. ( 2001 ). The LSW model for domain coarsening: Asymptotic behavior for total conserved mass . J. Stat. Phys. 104 ( 5/6 ): 1113 – 1144 . [CSA] [CROSSREF] [Crossref], [Web of Science ®] , [Google Scholar]) a necessary criterion for convergence to any self-similar solution which behaves like a finite power at the end of its (compact) support is given. It says that the data have to be regularly varying at the end of their support with the same power. This criterion is also shown to be sufficient if the power is sufficiently small and for data which are close to self-similar.

In this article we extend the local stability result to the whole range of self-similar solutions with compact support. Our first main result establishes global stability of self-similar solutions with not too large power. The proof relies on a global contraction argument for the spreading of characteristics. In addition, we also establish upper and lower bounds for the coarsening rates of the system for a suitable class of initial data whose variation is bounded at the end of the support but not necessarily regular.     

Penalized Orthogonal Iteration for Sparse Estimation of Generalized Eigenvalue Problem

We propose a new algorithm for sparse estimation of eigenvectors in generalized eigenvalue problems (GEPs). The GEP arises in a number of modern data-analytic situations and statistical methods, including principal component analysis (PCA), multiclass linear discriminant analysis (LDA), canonical correlation analysis (CCA), sufficient dimension reduction (SDR), and invariant co-ordinate selection. We propose to modify the standard generalized orthogonal iteration with a sparsity-inducing penalty for the eigenvectors. To achieve this goal, we generalize the equation-solving step of orthogonal iteration to a penalized convex optimization problem. The resulting algorithm, called penalized orthogonal iteration, provides accurate estimation of the true eigenspace, when it is sparse. Also proposed is a computationally more efficient alternative, which works well for PCA and LDA problems. Numerical studies reveal that the proposed algorithms are competitive, and that our tuning procedure works well. We demonstrate applications of the proposed algorithm to obtain sparse estimates for PCA, multiclass LDA, CCA, and SDR. Supplementary materials for this article are available online.     

Validity of the One and One-Half Layer Quasi-Geostrophic Model and Effective Topography

ABSTRACT

We rigorously justify the so-called one and one-half layer quasi-geostrophic model from the two layer model as the ratio of the depth of the bottom layer over that of the top layer approaches infinity. The effective dynamics is given by the classical barotropic quasi-geostrophic dynamics for the bottom layer without topography, and the one layer quasi-geostrophic dynamics with the stream function of the bottom layer serving as an effective (possibly time-dependent) topography for the the top layer. Such a one and one-half layer model is utilized in successful quantitative prediction of the Great Red Spot on Jupiter (see Turkington et al., 2001 Turkington , B. , Majda , A. J. , Haven , K. , DiBattista , M. ( 2001 ). Statistical equilibrium predictions of jets and spots on jupiter. Proc. Natl. Acad. Sci. USA 98 ( 22 ): 12346 – 12350 . [PUBMED] [INFOTRIEVE] [CSA] [CROSSREF] [Crossref] , [Google Scholar]).     

Weighted Solyanik Estimates for the Hardy–Littlewood Maximal Operator and Embedding of into

Let Open image in new window denote a weight in Open image in new window which belongs to the Muckenhoupt class Open image in new window and let Open image in new window denote the uncentered Hardy–Littlewood maximal operator defined with respect to the measure Open image in new window . The sharp Tauberian constant of Open image in new window with respect to Open image in new window , denoted by Open image in new window , is defined by

Open image in new window

In this paper, we show that the Solyanik estimate

$$\begin{aligned} \lim _{\alpha \rightarrow 1^-}\mathsf{C}_{w}(\alpha ) = 1 \end{aligned}$$

holds. Following the classical theme of weighted norm inequalities we also consider the sharp Tauberian constants defined with respect to the usual uncentered Hardy–Littlewood maximal operator Open image in new window and a weight Open image in new window :

Open image in new window

We show that we have Open image in new window if and only if Open image in new window . As a corollary of our methods we obtain a quantitative embedding of Open image in new window into Open image in new window .

     

Note on Conic Bundles over Henselian Discrete Valued Fields with Real Closed Residue Field

ABSTRACT

Let ? be a complete set of Sylow subgroups of a finite group G, that is, ? contains exactly one and only one Sylow p-subgroup of G for each prime p. A subgroup H of a finite group G is said to be ?-permutable if H permutes with every member of ?. The purpose of this article is to study the influence of ?-permutability of all maximal subgroups of the Sylow subgroups of the generalized Fitting subgroup of some normal subgroup of a finite group G on the structure of G. Our results improve and extend the main results of Asaad (1998 Asaad , M. ( 1998 ). On maximal subgroups of Sylow subgroups of finite groups . Comm. Algebra 26 ( 11 ): 3647 – 3652 . [CSA] [Taylor &; Francis Online], [Web of Science ®] , [Google Scholar]), Asaad and Heliel (2003 Asaad , M. , Heliel , A. A. ( 2003 ). On permutable subgroups of finite groups . Arch. Math. 80 : 113 – 118 . [CROSSREF] [CSA] [Crossref], [Web of Science ®] , [Google Scholar]), Asaad et al. (1991 Asaad , M. , Ramadan , M. , Shaalan , A. ( 1991 ). Influence of π-quasinormality on maximal subgroups of Sylow subgroups of Fitting subgroup of a finite group . Arch. Math. 56 : 521 – 527 . [CROSSREF] [CSA] [Crossref], [Web of Science ®] , [Google Scholar]), Li et al. (2003 Li , Y. , Wang , Y. , Wei , H. ( 2003 ). The influence of π-quasinormality of maximal subgroups of Sylow subgroups of a finite group . Arch. Math. 81 ( 3 ): 245 – 252 . [CROSSREF] [CSA] [Crossref], [Web of Science ®] , [Google Scholar]), Ramadan (1992 Ramadan , M. ( 1992 ). Influence of normality on maximal subgroups of Sylow subgroups of a finite group . Acta Math. Hungar. 59 ( 1–2 ): 107 – 110 . [CSA] [Crossref], [Web of Science ®] , [Google Scholar]), and Srinivasan (1980 Srinivasan , S. ( 1980 ). Two sufficient conditions for supersolvability of finite groups . Israel J. Math. 35 : 210 – 214 . [CSA] [Crossref], [Web of Science ®] , [Google Scholar]).     

Abelian groups small with respect to reduced groups

In this paperwe study

Open image in new window

-small abelian groups, where

Open image in new window

is the class of all reduced groups.

     

Positive definite and Gram tensor complementarity problems

Given a continuous function Open image in new window and Open image in new window , the non-linear complementarity problem \(\text{ NCP }(g,q)\) is to find a vector Open image in new window such that

$$\begin{aligned} x \ge 0,~~y:=g(x) +q\ge 0~~\text{ and }~~x^Ty=0. \end{aligned}$$

We say that g has the Globally Uniquely Solvable (\(\text{ GUS }\))-property if \(\text{ NCP }(g,q)\) has a unique solution for all Open image in new window and C-property if \(\mathrm{NCP}(g,q)\) has a convex solution set for all Open image in new window . In this paper, we find a class of non-linear functions that have the \(\text{ GUS }\)-property and C-property. These functions are constructed by some special tensors which are positive semidefinite. We call these tensors as Gram tensors.

     

Strongly Gauduchon Spaces

The authors define strongly Gauduchon spaces and the class■■, which are generalization of strongly Gauduchon manifolds in complex spaces. Comparing with the case of Kahlerian, the strongly Gauduchon space and the class■are similar to the Kahler space and the Fujiki class■■ respectively. Some properties about these complex spaces are obtained. Moreover, the relations between the strongly Gauduchon spaces and the class■■ are studied.     

Statistical simulation of one type of pairs of random variables with the use of fictitious jumps

Statistical simulation based on the sampling technique for a pair of random variables (T,

Open image in new window

), where T ∈ [0, +∞) and

Open image in new window

∈ ? ^d (d ≥ 1) is considered. The simultaneous distribution of the pair is specified in the form that is common for analogous problems in various fields. It has the form

Open image in new window

, where f is a function and m is a measure. The first variable T is the well-known random waiting time. A simulation method for the pair (T,

Open image in new window

) is constructed using a realization of an auxiliary Markov sequence of trial pairs. Applications of this method in particle transport theory and in kinetics of rarefied gases are discussed.