Showing papers in "Computational Statistics & Data Analysis in 2017"

TL;DR: In this paper, a new semiparametric quantile regression method is introduced based on sequentially fitting a likelihood optimal D-vine copula to given data resulting in highly flexible models with easily extractable conditional quantiles.

TL;DR: RHSBoost appears to be an attractive classification model for imbalance data and uses random undersampling and ROSE sampling under a boosting scheme to address the imbalance classification problem.

TL;DR: In this paper, the robust and asymptotic properties of 1-penalized MM-estimators and MM estimators with an adaptive 1 penalty are studied for the case of a fixed number of covariates.

TL;DR: A novel randomised response-adaptive design is proposed which maximises the total number of patient successes in the trial and penalises if a minimum number of patients are not recruited to each treatment arm.

TL;DR: This article introduces a bivariate copula additive model with continuous margins for location, scale and shape that permits the copula dependence and marginal distribution parameters to be estimated simultaneously and, like in GAMLSS, each parameter to be modeled using an additive predictor.

TL;DR: The Gaussian Copula method as discussed by the authors uses a 2-dimensional Gaussian copula to estimate the bivariate posterior for each pair of parameters separately, and then combines these estimates together to obtain the joint posterior.

TL;DR: A fast algorithm for computing the two-sample KolmogorovSmirnov test statistic is proposed, which is O(n) times more efficient than the brute force algorithm, where n is the sum of the two sample sizes.

TL;DR: It is demonstrated that the proposed corrections successfully remove the rare events bias and outperform the other ensemble classifiers that were considered and large flexibility and high interpretability of the proposed methods is also illustrated.

TL;DR: Several nonparametric estimators outperform commonly used treatment estimators based on parametric propensity scores in terms of root mean squared error (RMSE), even though average RMSEs based on the 16 simulation designs considered are not statistically significantly different across the estimators investigated.

TL;DR: In this paper, an estimator of the eigenvalues of the population covariance matrix has been proposed that is consistent according to a mean-squared criterion under large-dimensional asymptotics.

TL;DR: In this paper, a general approach for variable selection with shrinkage priors is proposed, which can be used along with any shrinkage prior, and has good performance in a wide range of synthetic data examples and in a real data example on selecting genes affecting survival due to lymphoma.

TL;DR: Two greedy algorithms for automatically selecting vine structures and component pair-copula building blocks are introduced and outperforms a Gaussian copula benchmark using both in-sample and out-of-sample criteria.

TL;DR: Two nonparametric estimators, which are based on the Beran estimator of the conditional survival function, are proved to be the local maximum likelihood estimators for mixture cure models.

TL;DR: This work introduces statistics that provably estimate the distance and direction of the boundary, which allows for a cut-and-normalize boundary correction, and introduces a consistent kernel density estimator that has uniform bias, at interior and boundary points, on manifolds with boundary.

TL;DR: A novel feature screening procedure is proposed for ultrahigh-dimensional survival data which is invariant to the monotone transformation of the response and can be readily applied to ultra high-dimensional complete data when the censoring rate is zero.

TL;DR: A class of multivariate linear models under the longitudinal setting, in which unobserved heterogeneity may evolve over time, is introduced, and a latent structure is considered to model heterogeneity, having a discrete support and following a first-order Markov chain.

TL;DR: The developed methodology is applied to estimate the proportion of people under the poverty line by counties and sex in Galicia (a region in north-west of Spain).

TL;DR: This paper focuses on exponential-family models for dependent data, which have applications in a wide variety of areas, but the dependence often results in an intractable likelihood, requiring either analytic approximation or MCMC-based techniques to fit.

TL;DR: Several equivariant estimators of multivariate location and scatter are studied, which are highly robust, have a controllable finite-sample efficiency and are computationally feasible in large dimensions.

TL;DR: A Smoothly Clipped Absolute Deviation penalization of the likelihood is proposed to shrink the parameters towards zeros and regularize the inference problem which is generally ill-posed.

TL;DR: A model free feature screening procedure based on the inverse probability weighted methods has been proposed, where the Kolmogorov filter method is used to screen the important features under an unknown propensity score function.

TL;DR: In this article, a more general solution is presented where the above mentioned limitation is relaxed and the presented solution can be easily adopted also for the task of efficient computation of integrated density derivative functionals involving an arbitrary derivative order.

TL;DR: The proposed Bayesian method is extended to multidimensional predictors such that the quantile regression depends on the predictors through an unknown linear combination only.

TL;DR: In this paper, an efficient sequential-minimal-optimization-based solver is developed and its convergence derived for the underlying optimization problem, and the results are compared with the solver for quantile regression and the recent R-package ER-Boost.

TL;DR: It is shown that the JPS scheme improves estimation of the population proportion in a very wide range of settings as compared to simple random sampling (SRS).

TL;DR: It is shown that a specific wild bootstrap procedure inherits the large sample properties of the Wald- and ANOVA-type statistics while considerably improving their small sample behavior.

TL;DR: It is shown that a subclass of Lq penalties with q less than or equal to one can be expressed as sums of L2 penalties, and it follows that the lasso and other norm-penalized regression estimates may be obtained using a very simple and intuitive alternating ridge regression algorithm.

TL;DR: A frailty model-based maximum likelihood approach is proposed with the use of monotone splines to approximate the unknown baseline cumulative hazard function of the failure time and a novel EM algorithm, which is based on a three-stage data augmentation and can be easily implemented, is presented.

TL;DR: A population-based evolutionary algorithm called imperialist competitive algorithm (ICA) is applied to find minimax or nearly minimax D-optimal designs for nonlinear models and can hybridize with a local search to find optimal designs under a more complicated criterion, such as standardized maximin optimality.

TL;DR: A constrained center and range joint model to fit linear regression to interval-valued symbolic data is introduced that has better fitness and avoids the negative value of the range of the predicted dependent interval variable by adding nonnegative constraints.