Stochastic comparisons on conditional residual lifetime and inactivity time of coherent systems with exchangeable components
Publication date: Available online 16 October 2018Source: Statistics & Probability LettersAuthor(s): Ebrahim Salehi, Mahdi TavangarAbstractThe paper describes and develops two concepts of conditional residual lifetime of the system, given that at a fixed time a certain number of components have failed but still there are some functioning components, and conditional inactivity time of the failed components in the system, given that at that time a certain number of components have failed but the system is still alive. We consider the m-out-of-n andthe coherent systems consisting of n components whose lifetimes are assume...
Source: Statistics and Probability Letters - October 17, 2018 Category: Statistics Source Type: research

On LR simultaneous test of high-dimensional mean vector and covariance matrix under non-normality
Publication date: Available online 16 October 2018Source: Statistics & Probability LettersAuthor(s): Zhenzhen Niu, Jiang Hu, Zhidong Bai, Wei GaoAbstractIn this paper, we primarily focus on simultaneous testing mean vector and covariance matrix with high-dimensional non-Gaussian data, based on the classical likelihood ratio test. Applying the central limit theorem for linear spectral statistics of sample covariance matrices, we establish new modification for the likelihood ratio test, and find that this modified test converges in distribution to normal distribution, when the dimension p tends to infinity, proportionate...
Source: Statistics and Probability Letters - October 17, 2018 Category: Statistics Source Type: research

A model for level induced conditional heteroskedasticity
Publication date: Available online 15 October 2018Source: Statistics & Probability LettersAuthor(s): Jon Michel, Robert M. de JongAbstractA class of conditional heteroskedasticity models is introduced and analyzed. This class of models is motivated by the desire to allow the level of a GARCH process to influence the volatility. We show the existence of a unique strictly stationary solution which is β-mixing. The analysis of this model does not rely upon Markov chain methods. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 16, 2018 Category: Statistics Source Type: research

Minimum distance parameter estimation for SDEs with small α-stable noises
Publication date: Available online 15 October 2018Source: Statistics & Probability LettersAuthor(s): Huiyan Zhao, Chongqi ZhangAbstractWe consider the minimum distance estimate for stochastic nonlinear differential equations with small α-stable noises, where α∈(0,2). The consistency property and limit distribution are studied for fixed time as the diffusion coefficient goes to zero. Moreover, the 1-dimensional case of asymptotic law of the limit distribution is considered for large time. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 16, 2018 Category: Statistics Source Type: research

Probabilistic interpretation for Sobolev solutions of McKean-Vlasov partial differential equations
Publication date: Available online 15 October 2018Source: Statistics & Probability LettersAuthor(s): Zhen Wu, Ruimin XuAbstractIn this paper, we give a probabilistic interpretation of Sobolev solutions to parabolic semilinear McKean-Vlasov partial differential equations (PDEs for short) in terms of mean-field backward stochastic differential equations (BSDEs for short). This probabilistic interpretation can be viewed as a generalization of the Feynman–Kac formula. The method is based on the stochastic flow technique which is different from classical stochastic differential equations (SDEs for short) due to the in...
Source: Statistics and Probability Letters - October 16, 2018 Category: Statistics Source Type: research

Bayesian estimation of Kendall’s τ using a latent normal approach
Publication date: Available online 15 October 2018Source: Statistics & Probability LettersAuthor(s): Johnny van Doorn, Alexander Ly, Maarten Marsman, Eric-Jan WagenmakersAbstractThe rank-based association between two variables can be modeled by introducing a latent normal level to ordinal data. We demonstrate how this approach yields Bayesian inference for Kendall’s τ, improving on a recent Bayesian solution based on its asymptotic properties. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 16, 2018 Category: Statistics Source Type: research

A variation of constant formula for Caputo fractional stochastic differential equations
Publication date: Available online 15 October 2018Source: Statistics & Probability LettersAuthor(s): P.T. Anh, T.S. Doan, P.T. HuongAbstractWe establish and prove a variation of constant formula for Caputo fractional stochastic differential equations whose coefficients satisfy a standard Lipschitz condition. The main ingredient in the proof is to use Ito’s representation theorem and the known variation of constant formula for deterministic Caputo fractional differential equations. As a consequence, for these systems we point out the coincidence between the notion of classical solutions introduced in Wang et ...
Source: Statistics and Probability Letters - October 16, 2018 Category: Statistics Source Type: research

Stable Lévy process delayed by tempered stable subordinator
Publication date: Available online 12 October 2018Source: Statistics & Probability LettersAuthor(s): J. Gajda, A. Kumar, A. WyłomańskaAbstractWe consider symmetric stable Lévy motion time-changed by tempered stable subordinator. This process generalizes the normal inverse Gaussian process without drift term, introduced by Barndorff-Nielsen. The asymptotic tail behavior of the density function of this process and corresponding Lévy density is obtained. The governing Fokker–Planck–Kolmogorov equation of the density function of the introduced process in terms of shifted fractional derivative is...
Source: Statistics and Probability Letters - October 13, 2018 Category: Statistics Source Type: research

Conditional distributions of multivariate normal mean–variance mixtures
Publication date: Available online 13 October 2018Source: Statistics & Probability LettersAuthor(s): Ahad Jamalizadeh, Narayanaswamy BalakrishnanAbstractIn this paper, we show that the conditional distribution of a multivariate normal mean–variance mixture (MNMVM) distribution is also a MNMVM distribution.We show the usefulness of the established result by deriving the conditional distributions of multivariate generalized hyperbolic distribution. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 13, 2018 Category: Statistics Source Type: research

Moments and cumulants of the extremes of a sample from a uniform distribution
Publication date: Available online 11 October 2018Source: Statistics & Probability LettersAuthor(s): Christopher S. Withers, Saralees NadarajahAbstractWe give expansions in inverse powers of the sample size n for the joint moments and cumulants of the extremes when sampling from a uniform distribution. Estimates of low bias are given for smooth functions of the end points and their performance illustrated by simulations. Some extensions are given for general bivariate extreme estimates. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 12, 2018 Category: Statistics Source Type: research

Central limit theorem for the entries of products of random matrices without the positivity condition
Publication date: Available online 10 October 2018Source: Statistics & Probability LettersAuthor(s): Ignacio ArbuésAbstractWe prove a Central Limit Theorem for the entries of products of i.i.d. random matrices, in which the factors may have both positive and negative elements. We focus in the case that the distribution of the matrices has a density. The theorem is proved by representing the products as Markov Chains and establishing a variational inequality for a certain Lyapunov function. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 11, 2018 Category: Statistics Source Type: research

Integration by parts formulas for marked Hawkes processes
Publication date: Available online 11 October 2018Source: Statistics & Probability LettersAuthor(s): Atsushi TakeuchiAbstractConsider the processes {Lt;t≥0} and {λt;t≥0} associated with a marked Hawkes process and its conditional intensity. In the present paper, the integration by parts formulas for those processes are constructed, and applied to the study on the absolute continuity for their conditional laws. Moreover, the sensitivity formulas for those processes with respect to the parameters are given. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 11, 2018 Category: Statistics Source Type: research

Concentration of measure for radial distributions and consequences for statistical modeling
Publication date: Available online 9 October 2018Source: Statistics & Probability LettersAuthor(s): Ery Arias-Castro, Xiao PuAbstractMotivated by problems in high-dimensional statistics such as mixture modeling for classification and clustering, we consider the behavior of radial densities as the dimension increases. We establish a form of concentration of measure, and even a convergence in distribution, under additional assumptions. This extends the well-known behavior of the normal distribution (its concentration around the sphere of radius square-root of the dimension) to other radial densities. We draw some possibl...
Source: Statistics and Probability Letters - October 10, 2018 Category: Statistics Source Type: research

A note on strong-consistency of componentwise ARH(1) predictors
Publication date: Available online 9 October 2018Source: Statistics & Probability LettersAuthor(s): M.D. Ruiz-Medina, J. Álvarez-LiébanaAbstractNew results on strong-consistency in the trace operator norm are obtained, in the parameter estimation of an autoregressive Hilbertian process of order one (ARH(1) process). Additionally, a strongly-consistent diagonal componentwise estimator of the autocorrelation operator is derived, based on its empirical singular value decomposition. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 10, 2018 Category: Statistics Source Type: research

Partial monotonicity of entropy revisited
Publication date: Available online 9 October 2018Source: Statistics & Probability LettersAuthor(s): Wanwan XiaAbstractIn this short note, we revisit the partial monotonicity of the differential entropy, and point out some errors in Shangari and Chen (2012) and Gupta and Bajaj (2013). The partial monotonicity of the Shannon entropy is also established. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 10, 2018 Category: Statistics Source Type: research

Time-varying cointegration model using wavelets
Publication date: Available online 9 October 2018Source: Statistics & Probability LettersAuthor(s): Eder Lucio da Fonseca, Airlane Pereira Alencar, Pedro Alberto MorettinAbstractThis work proposes a wavelet based Vector Error Correction Model with time-varying cointegration. The maximum likelihood estimators and likelihood ratio statistics were evaluated based on Monte Carlo and Bootstrap simulations. The model was used to study the Purchasing Power Parity hypothesis. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 10, 2018 Category: Statistics Source Type: research

Another bias correction for asymmetric kernel density estimation with a parametric start
Publication date: Available online 24 September 2018Source: Statistics & Probability LettersAuthor(s): Masayuki Hirukawa, Mari SakudoAbstractThis paper studies yet another semiparametric bias-corrected density estimation using asymmetric kernels. The estimator can be obtained by making a multiplicative bias correction for the initial parametric model twice, and it is shown to establish rate improvement when best implemented. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

On three-level D-optimal paired choice designs
Publication date: Available online 24 September 2018Source: Statistics & Probability LettersAuthor(s): Rakhi SinghAbstractConsidering three-level D-optimal paired choice designs for estimating all the main effects and two-factor interaction effects under the utility-neutral multinomial logit model, we provide general characteristics (and examples) required in generators allowing significant reduction in the number of choice pairs. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

An extremal property of the normal distribution, with a discrete analog
Publication date: Available online 25 September 2018Source: Statistics & Probability LettersAuthor(s): Erwan Hillion, Oliver Johnson, Adrien SaumardAbstractWe give a new proof, using only the Brascamp–Lieb inequality, of the fact that the Gaussian measure is the only strong log-concave measure having a strong log-concavity parameter equal to its covariance matrix. We unify the continuous and discrete settings by also giving a similar characterization of the Poisson measure in the discrete case, using “Chebyshev’s other inequality”. We briefly discuss how these results relate to Stein and Stein&n...
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

A note on Parisian ruin under a hybrid observation scheme
Publication date: Available online 26 September 2018Source: Statistics & Probability LettersAuthor(s): Mohamed Amine LkabousAbstractIn this paper, we study the concept of Parisian ruin under the hybrid observation scheme model introduced by Li et al. (2018). Under this model, the process is observed at Poisson arrival times whenever the business is financially healthy and it is continuously observed when it goes below 0. The Parisian ruin is then declared when the process stays below zero for a consecutive period of time greater than a fixed delay. We improve the result originally obtained in Li et al. (2018)...
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Bounds for the probability to leave the interval
Publication date: Available online 26 September 2018Source: Statistics & Probability LettersAuthor(s): V.I. LotovAbstractWe obtain upper and lower bounds for the probability that random walk leaves the strip through the upper boundary. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Asymptotic normality of kernel density function estimator from continuous time stationary and dependent processes
Publication date: Available online 1 October 2018Source: Statistics & Probability LettersAuthor(s): Naâmane Laïb, Djamal LouaniAbstractOur purpose in this work is to establish the asymptotic normality for the kernel density function estimator in the setting of continuous time stationary and dependent data. Our results allow to construct confidence bands for the density f(x). The proof techniques use martingale difference devices and a sequence of projections on appropriate σ-fields. A numerical study is performed to illustrate the impact of processes sampling. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Geometric dispersion models with real quadratic v-functions
Publication date: Available online 1 October 2018Source: Statistics & Probability LettersAuthor(s): Rahma Abid, Célestin C. Kokonendji, Afif MasmoudiAbstractGeometric dispersion models, characterized by their v-functions, are recently introduced arising from geometric sums of exponential dispersion models and they exhibit many potential applications. In this paper, we classify all the real quadratic v-functions. Up to affinity, there are only six types of such models with unbounded domain: asymmetric Laplace, geometric and the remaining four are obtained by the exponential mixtures of Poisson, gamma, negative bi...
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Stable limit theory for the Gaussian QMLE in a non-stationary asymmetric GARCH model
Publication date: Available online 3 October 2018Source: Statistics & Probability LettersAuthor(s): Stelios ArvanitisAbstractWe derive the limit theory of the Gaussian QMLE in a non-stationary Asymmetric GARCH(1,1) model when the squared innovation process lies in the domain of attraction of a stable law. When the stability parameter lies in 1,2, we find regularly varying rates and stable limits for the QMLE of the asymmetry and GARCH parameters. When it is less than one we derive total inconsistency. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Kernel estimation of the conditional density under a censorship model
Publication date: Available online 3 October 2018Source: Statistics & Probability LettersAuthor(s): Lamia Aouicha, Fatiha MessaciAbstractWe establish the mean square convergence, with rate, for an introduced kernel estimator of the conditional density function when the response variable is twice censored. The common case of right censored data can be derived as a particular case. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - October 4, 2018 Category: Statistics Source Type: research

Tamed Euler–Maruyama approximation for stochastic differential equations with locally Hölder continuous diffusion coefficients
Publication date: Available online 21 September 2018Source: Statistics & Probability LettersAuthor(s): Hoang Long Ngo, Duc Trong LuongAbstractWe study the strong convergence of the tamed Euler–Maruyama approximation for stochastic differential equations whose coefficients are superlinearly grow and diffusion coefficient is locally Hölder continuous. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 22, 2018 Category: Statistics Source Type: research

Stein operators for variables form the third and fourth Wiener chaoses
Publication date: Available online 14 September 2018Source: Statistics & Probability LettersAuthor(s): Robert E. GauntAbstractLet Z be a standard normal random variable and let Hn denote the nth Hermite polynomial. In this note, we obtain Stein equations for the random variables H3(Z) andH4(Z), which represents a first step towards developing Stein’s method for distributional limits from the third and fourth Wiener chaoses. Perhaps surprisingly, these Stein equations are fifth and third order linear ordinary differential equations, respectively. As a warm up, we obtain a Stein equation for the random variable aZ2...
Source: Statistics and Probability Letters - September 15, 2018 Category: Statistics Source Type: research

Editorial Board
Publication date: December 2018Source: Statistics & Probability Letters, Volume 143Author(s): (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 15, 2018 Category: Statistics Source Type: research

Efficient estimation of a distribution based on censored data
Publication date: Available online 14 September 2018Source: Statistics & Probability LettersAuthor(s): Filippos Alevizos, Dimitrios Bagkavos, Dimitrios IoannidesAbstractThe asymptotic relative deficiency of the Kaplan –Meier over an existing kernel estimate is established. Further, finite sample numerical evidence is provided suggesting that the kernel estimate is preferable when performance is measured through the mean square error. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 14, 2018 Category: Statistics Source Type: research

On the semi-group of a scaled skew Bessel process
Publication date: Available online 8 September 2018Source: Statistics & Probability LettersAuthor(s): Larbi Alili, Andrew AylwinAbstractWe define scaled skew Bessel processes and determine their Green’s functions and semi-group densities. We prove that these processes satisfy the time inversion property although the corresponding densities of the semi-groups are not, in general, twice differentiable in the density and starting point arguments. Some characterizations and semi-martingale properties are given. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 9, 2018 Category: Statistics Source Type: research

Lift expectations of random sets
Publication date: Available online 8 September 2018Source: Statistics & Probability LettersAuthor(s): Marc-Arthur Diaye, Gleb A. Koshevoy, Ilya MolchanovAbstractIt is known that the distribution of an integrable random vector ξ in Rd is uniquely determined by a (d+1)-dimensional convex body called the lift zonoid of ξ. This concept is generalised to define the lift expectation of random convex bodies. However, the unique identification property of distributions is lost; it is shown that the lift expectation uniquely identifies only one-dimensional distributions of the support function, and so different random con...
Source: Statistics and Probability Letters - September 9, 2018 Category: Statistics Source Type: research

Estimation of minimum and maximum correlation coefficients
Publication date: Available online 7 September 2018Source: Statistics & Probability LettersAuthor(s): Jim X. XiangAbstractTo generate correlated data for given marginal distributions, it is essential that the desired Pearson correlation coefficient is between the minimum and maximum correlation coefficients. In this paper, we consider estimation of the minimum and maximum correlation coefficients of continuous random variables X andY. A strong law of large numbers and asymptotic normality are established for the estimators studied in this paper. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 8, 2018 Category: Statistics Source Type: research

Linear process bootstrap unit root test
Publication date: Available online 7 September 2018Source: Statistics & Probability LettersAuthor(s): Nan Zou, Dimitris N. PolitisAbstractOne of the most widely applied unit root tests suffers from size distortions when moving average noise exists. As a remedy, this paper proposes a bootstrap test targeting moving average noise and shows its effectiveness in both theory and simulation. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 8, 2018 Category: Statistics Source Type: research

Three skewed matrix variate distributions
Publication date: Available online 6 September 2018Source: Statistics & Probability LettersAuthor(s): Michael P.B. Gallaugher, Paul D. McNicholasAbstractThree-way data can be conveniently modelled by using matrix variate distributions. Although there has been a lot of work for the matrix variate normal distribution, there is little work in the area of matrix skew distributions. Three matrix variate distributions that incorporate skewness, as well as other flexible properties such as concentration, are discussed. Equivalences to multivariate analogues are presented, and moment generating functions are derived. Maximum l...
Source: Statistics and Probability Letters - September 7, 2018 Category: Statistics Source Type: research

On probabilistic aspects of Chebyshev polynomials
Publication date: Available online 6 September 2018Source: Statistics & Probability LettersAuthor(s): Paweł J. SzabłowskiAbstractThe main goal of this note is to provide new, mostly multidimensional densities, compactly supported and list many of its properties that enable effective calculations. The idea of obtaining such densities is firstly to build some one-dimensional densities depending on many parameters and then treat the constructed in this way distributions as conditional ones. Then of course by imposing certain distribution on the parameters and multiplying the two distributions we arrive at new multivaria...
Source: Statistics and Probability Letters - September 7, 2018 Category: Statistics Source Type: research

Efficient computation for Bayesian comparison of two proportions
Publication date: Available online 4 September 2018Source: Statistics & Probability LettersAuthor(s): Mikkel N. Schmidt, Morten MørupAbstractIn Bayesian comparison of two proportions, the exact computation of the evidence involves evaluating a generalized hypergeometric function. Several agreeing, but not identical, expressions for the evidence have been derived in the literature; however, their practical computation (by summing the truncated hypergeometric series) can be troubled by slow convergence or catastrophic cancellation. Using a set of equivalence relations for the generalized hypergeometric function, w...
Source: Statistics and Probability Letters - September 5, 2018 Category: Statistics Source Type: research

On asymptotic normality of certain linear rank statistics
Publication date: Available online 31 August 2018Source: Statistics & Probability LettersAuthor(s): V. SkorniakovAbstractWe consider asymptotic normality of linear rank statistics under various randomization rules met in clinical trials. Exposition relies on some general limit theorem given in McLeish (1974). Examples of applications include well known results as well as new ones. Though exposition is tightly attached to the context of clinical trials, the obtained results equally well apply to situations where randomization of similar kind takes place. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 1, 2018 Category: Statistics Source Type: research

Large sample properties of a new measure of income inequality
Publication date: Available online 1 September 2018Source: Statistics & Probability LettersAuthor(s): Xiaoke Zhang, Joseph L. GastwirthAbstractThis paper develops the asymptotic properties of a recently proposed measure of inequality emphasizing the status of the poor relative to the rich. A simulation confirms the asymptotic results are applicable in practice. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - September 1, 2018 Category: Statistics Source Type: research

Using DAGs to identify the sufficient dimension reduction in the Principal Fitted Components model
Publication date: Available online 27 August 2018Source: Statistics & Probability LettersAuthor(s): María Eugenia Szretter NosteAbstractWe identify the sufficient reduction for the Principal Fitted Components model under mild conditions which generalize those considered in previous works. We give a short proof of the main result based on the d-separation for directed acyclic graphs (DAGs), linking two areas that, to our knowledge, have not been linked before in statistics. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 28, 2018 Category: Statistics Source Type: research

Exponential probability distribution on symmetric matrices
Publication date: Available online 27 August 2018Source: Statistics & Probability LettersAuthor(s): A. Hassairi, A. RoulaAbstractIn this paper, we define in the most natural way a multivariate extension of the exponential distribution as a particular Wishart on the cone of positive definite symmetric matrices. We also introduce a notion of reliability function for a matrix random variable. We then show that, under a condition of invariance, the exponential distribution on symmetric matrices is characterized by a property of memoryless generalizing and including the characterization established for the real exponential ...
Source: Statistics and Probability Letters - August 28, 2018 Category: Statistics Source Type: research

The scale enhanced wild bootstrap method for evaluating climate models using wavelets
Publication date: Available online 23 August 2018Source: Statistics & Probability LettersAuthor(s): Snigdhansu ChatterjeeAbstractA novel resampling-based method is presented for testing if a Physics-based model or reanalysis object successfully emulates climate signals. We establish asymptotic consistency, present simulations and a real-data application to illustrate the performance of the proposed method. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 24, 2018 Category: Statistics Source Type: research

Correlation extrapolated
Publication date: Available online 23 August 2018Source: Statistics & Probability LettersAuthor(s): Pierre-André G. MaugisAbstractWe provide consistent and asymptotic normal estimators of correlation that corrects the bias induced by partial samples. Through examples, we show that the estimators behaves well in small-sample and yields powerful methodologies for non-linear regression as well as dependence testing. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 24, 2018 Category: Statistics Source Type: research

On APF test for Poisson process with shift and scale parameters
Publication date: Available online 20 August 2018Source: Statistics & Probability LettersAuthor(s): A.S. Dabye, Yu. A. Kutoyants, E.D. TanguepAbstractWe propose the goodness of fit test for inhomogeneous Poisson processes with unknown scale and shift parameters. A test statistic of Cramér-von Mises type is proposed and its asymptotic behavior is studied. We show that under null hypothesis the limit distribution of this statistic does not depend on unknown parameters. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 21, 2018 Category: Statistics Source Type: research

Robust conditional nonparametric independence screening for ultrahigh-dimensional data
This article novelly proposes a robust model-free screening procedure, which performs well for a variety of semivarying coefficient models. Under technical conditions, we show that it possesses the ranking consistency property and the sure screening property. Comprehensive simulation studies are conducted to demonstrate that it exhibits more competitive performance than existing screening methods. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 17, 2018 Category: Statistics Source Type: research

On the consistency of penalized MLEs for Erlang mixtures
Publication date: Available online 16 August 2018Source: Statistics & Probability LettersAuthor(s): Cuihong Yin, X. Sheldon Lin, Rongtan Huang, Haili YuanAbstractIn Yin and Lin (2016), a new penalty, termed as iSCAD penalty, is proposed to obtain the maximum likelihood estimates (MLEs) of the weights and the common scale parameter of an Erlang mixture model. In that paper, it is shown through simulation studies and a real data application that the penalty provides an efficient way to determine the MLEs and the order of the mixture. In this paper, we provides a theoretical justification and show that the penalized maxim...
Source: Statistics and Probability Letters - August 17, 2018 Category: Statistics Source Type: research

Passive tracer in non-Markovian, Gaussian velocity field
Publication date: Available online 11 August 2018Source: Statistics & Probability LettersAuthor(s): Tymoteusz ChojeckiAbstractWe consider the trajectory of a tracer that is the solution of an ordinary differential equation Ẋ(t)=V(t,X(t)),X(0)=0, with the right hand side, that is a stationary, zero-mean, Gaussian vector field with incompressible realizations. It is known, see Fannjiang and Komorowski (1999), Carmona and Xu (1996) and Komorowski et al. (2012), that X(t)∕t converges in law, as t→+∞, to a normal, zero mean vector, provided that the field V(t,x) is Markovian and has the spectral gap prop...
Source: Statistics and Probability Letters - August 11, 2018 Category: Statistics Source Type: research

Editorial Board
Publication date: November 2018Source: Statistics & Probability Letters, Volume 142Author(s): (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 11, 2018 Category: Statistics Source Type: research

Factorizable non-atomic copulas
Publication date: Available online 10 August 2018Source: Statistics & Probability LettersAuthor(s): Tanes Printechapat, Songkiat SumetkijakanAbstractGeneralizing the notion of invariant sets by Darsow and Olsen, Sumetkijakan studied a subclass of singular copulas, the so-called non-atomic copulas, defined via its associated σ-algebras. It was shown that the Markov operator of every non-atomic copula is partially factorizable, i.e. it is the composition of left and right invertible Markov operators on a subspace of L1([0,1]) depending on the copula. Here, we further investigate the associated σ-algebras...
Source: Statistics and Probability Letters - August 11, 2018 Category: Statistics Source Type: research

A Bayesian hierarchical model for multiple imputation of urban spatio-temporal groundwater levels
Publication date: Available online 8 August 2018Source: Statistics & Probability LettersAuthor(s): Kimberly F. Manago, Terri S. Hogue, Aaron Porter, Amanda S. HeringAbstractGroundwater levels in urban areas are irregularly sampled and not well understood. Using a separable space–time Bayesian Hierarchical Model, we obtain multiple imputations of the missing values to analyze spatial and temporal groundwater level fluctuations in Los Angeles, CA. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 9, 2018 Category: Statistics Source Type: research

Optimal directional statistic for general regression
Publication date: Available online 7 August 2018Source: Statistics & Probability LettersAuthor(s): Jonathan Gillard, Anatoly ZhigljavskyAbstractFor a general linear regression model we construct a directional statistic which maximizes the probability that the scalar product between the vector of unknown parameters and any linear estimator is positive. Special emphasis is given to comparison of this directional statistic with the BLUE and explaining why the BLUE could be relatively poor. We illustrate our results on analytical and numerical examples. (Source: Statistics and Probability Letters)
Source: Statistics and Probability Letters - August 8, 2018 Category: Statistics Source Type: research