Computational Statistics & Data Analysis最新文献

Modern network analysis often involves multi-layer network data in which the nodes are aligned, and the edges on each layer represent one of the multiple relations among the nodes. Current literature on multi-layer network data is mostly limited to undirected relations. However, direct relations are more common and may introduce extra information. This study focuses on community detection (or clustering) in multi-layer directed networks. To take into account the asymmetry, a novel spectral-co-clustering-based algorithm is developed to detect co-clusters, which capture the sending patterns and receiving patterns of nodes, respectively. Specifically, the eigendecomposition of the debiased sum of Gram matrices over the layer-wise adjacency matrices is computed, followed by the k-means, where the sum of Gram matrices is used to avoid possible cancellation of clusters caused by direct summation. Theoretical analysis of the algorithm under the multi-layer stochastic co-block model is provided, where the common assumption that the cluster number is coupled with the rank of the model is relaxed. After a systematic analysis of the eigenvectors of the population version algorithm, the misclassification rates are derived, which show that multi-layers would bring benefits to the clustering performance. The experimental results of simulated data corroborate the theoretical predictions, and the analysis of a real-world trade network dataset provides interpretable results.

This paper discusses regression analysis of case K interval-censored failure time data, a general type of failure time data, in the presence of informative censoring with the focus on simultaneous variable selection and estimation. Although many authors have considered the challenging variable selection problem for interval-censored data, most of the existing methods assume independent or non-informative censoring. More importantly, the existing methods that allow for informative censoring are frailty model-based approaches and cannot directly assess the degree of informative censoring among other shortcomings. To address these, we propose a conditional approach and develop a penalized sieve maximum likelihood procedure for the simultaneous variable selection and estimation of covariate effects. Furthermore, we establish the oracle property of the proposed method and illustrate the appropriateness and usefulness of the approach using a simulation study. Finally we apply the proposed method to a set of real data on Alzheimer's disease and provide some new insights.

Recent advances in DNA sequencing technology have led to a growing interest in microbiome data. Since the data are often high-dimensional, there is a clear need for dimensionality reduction. However, the compositional nature and zero-inflation of microbiome data present many challenges in developing new methodologies. New PCA methods for zero-inflated compositional data are presented, based on a novel framework called principal compositional subspace. These methods aim to identify both the principal compositional subspace and the corresponding principal scores that best approximate the given data, ensuring that their reconstruction remains within the compositional simplex. To this end, the constrained optimization problems are established and alternating minimization algorithms are provided to solve the problems. The theoretical properties of the principal compositional subspace, particularly focusing on its existence and consistency, are further investigated. Simulation studies have demonstrated that the methods achieve lower reconstruction errors than the existing log-ratio PCA in the presence of a linear pattern and have shown comparable performance in a curved pattern. The methods have been applied to four microbiome compositional datasets with excessive zeros, successfully recovering the underlying low-rank structure.

Bayesian inference procedures for the parameters of the multivariate random effects model are derived under the assumption of an elliptically contoured distribution when the Berger and Bernardo reference and the Jeffreys priors are assigned to the model parameters. A new numerical algorithm for drawing samples from the posterior distribution is developed, which is based on the hybrid Gibbs sampler. The new approach is compared to the two Metropolis-Hastings algorithms previously derived in the literature via an extensive simulation study. The findings are applied to a Bayesian multivariate meta-analysis, conducted using the results of ten studies on the effectiveness of a treatment for hypertension. The analysis investigates the treatment effects on systolic and diastolic blood pressure. The second empirical illustration deals with measurement data from the CCAUV.V-K1 key comparison, aiming to compare measurement results of sinusoidal linear accelerometers at four frequencies.

Tests of fit for classes of distributions that include the Weibull, the Pareto and the Fréchet families are proposed. The new tests employ the novel tool of the min–characteristic function and are based on an $L^2$–type weighted distance between this function and its empirical counterpart applied on suitably standardized data. If data–standardization is performed using the MLE of the distributional parameters then the method reduces to testing for the standard member of the family, with parameter values known and set equal to one. Asymptotic properties of the tests are investigated. A Monte Carlo study is presented that includes the new procedure as well as competitors for the purpose of specification testing with three extreme value distributions. The new tests are also applied on a few real–data sets.

High-dimensional accelerated failure time (AFT) models are commonly used regression models in survival analysis. Feature selection problem in high-dimensional AFT models is addressed, considering scenarios involving solely main effects or encompassing both main and interaction effects. A rank-based sequential feature selection (RankSFS) method is proposed, the selection consistency is established and illustrated by comparing it with existing methods through extensive numerical simulations. The results show that RankSFS achieves a higher Positive Discovery Rate (PDR) and lower False Discovery Rate (FDR). Additionally, RankSFS is applied to analyze the data on Breast Cancer Relapse. With a remarkable short computational time, RankSFS successfully identifies two crucial genes.

The effective control of infectious diseases relies on accurate assessment of the impact of interventions, which is often hindered by the complex dynamics of the spread of disease. A Beta-Dirichlet switching state-space transmission model is proposed to track underlying dynamics of disease and evaluate the effectiveness of interventions simultaneously. As time evolves, the switching mechanism introduced in the susceptible-exposed-infected-recovered (SEIR) model is able to capture the timing and magnitude of changes in the transmission rate due to the effectiveness of control measures. The implementation of this model is based on a particle Markov Chain Monte Carlo algorithm, which can estimate the time evolution of SEIR states, switching states, and high-dimensional parameters efficiently. The efficacy of the proposed model and estimation procedure are demonstrated through simulation studies. With a real-world application to British Columbia's COVID-19 outbreak, the proposed switching state-space transmission model quantifies the reduction of transmission rate following interventions. The proposed model provides a promising tool to inform public health policies aimed at studying the underlying dynamics and evaluating the effectiveness of interventions during the spread of the disease.

An empirical Bayes method for the Poisson matrix denoising and completion problems is proposed, and a corresponding algorithm called EBPM (Empirical Bayes Poisson Matrix) is developed. This approach is motivated by the non-central singular value shrinkage prior, which was used for the estimation of the mean matrix parameter of a matrix-variate normal distribution. Numerical experiments show that the EBPM algorithm outperforms the common nuclear norm penalized method in both matrix denoising and completion. The EBPM algorithm is highly efficient and does not require heuristic parameter tuning, as opposed to the nuclear norm penalized method, in which the regularization parameter should be selected. The EBPM algorithm also performs better than others in real-data applications.

A one-shot federated transfer learning method using random forests (FTRF) is developed to improve the prediction accuracy at a target data site by leveraging information from auxiliary sites. Both theoretical and numerical results show that the proposed federated transfer learning approach is at least as accurate as the model trained on the target data alone regardless of possible data heterogeneity, which includes imbalanced and non-IID data distributions across sites and model mis-specification. FTRF has the ability to evaluate the similarity between the target and auxiliary sites, enabling the target site to autonomously select more similar site information to enhance its predictive performance. To ensure communication efficiency, FTRF adopts the model averaging idea that requires a single round of communication between the target and the auxiliary sites. Only fitted models from auxiliary sites are sent to the target site. Unlike traditional model averaging, FTRF incorporates predicted outcomes from other sites and the original variables when estimating model averaging weights, resulting in a variable-dependent weighting to better utilize models from auxiliary sites to improve prediction. Five real-world data examples show that FTRF reduces the prediction error by 2-40% compared to methods not utilizing auxiliary information.

Linear log-contrast models have been widely used to describe the relationship between the response of interest and the compositional covariates, in which one central task is to identify the significant compositional covariates while controlling the false discovery rate (FDR) at a nominal level. To achieve this goal, a new FDR control method is proposed for linear log-contrast models with high-dimensional compositional covariates. An appealing feature of the proposed method is that it completely bypasses the traditional p-values and utilizes only the symmetry property of the test statistic for the unimportant compositional covariates to give an upper bound of the FDR. Under some regularity conditions, the FDR can be asymptotically controlled at the nominal level for the proposed method in theory, and the theoretical power is also proven to approach one as the sample size tends to infinity. The finite-sample performance of the proposed method is evaluated through extensive simulation studies, and applications to microbiome compositional datasets are also provided.