Showing papers in "Journal of Statistical Planning and Inference in 2022"

TL;DR: In this article, simultaneous confidence bands (SCBs) for functional parameters over arbitrary dimensional compact domains using the Gaussian Kinematic formula of t -processes (tGKF) were proposed.

TL;DR: In this paper, a normalized power prior approach which obeys the likelihood principle and is a modified form of the joint power prior was proposed, which minimizes the weighted Kullback-Leibler divergence.

TL;DR: In this paper , simultaneous confidence bands (SCBs) for functional parameters over arbitrary dimensional compact domains using the Gaussian Kinematic formula of t-processes (tGKF) were proposed.

TL;DR: In this paper , a normalized power prior approach which obeys the likelihood principle and is a modified form of the joint power prior was proposed, which minimizes the weighted Kullback-Leibler divergence.

TL;DR: In this article, the authors study the asymptotic efficiency of the calibration estimator with high-dimensional auxiliary data sets and prove that it may suffer from an additional variability that may not be neglected in certain conditions.

TL;DR: In this article , the asymptotic efficiency of the calibration estimator with high-dimensional auxiliary data sets is studied and it is shown that it may suffer from an additional variability that may not be neglected in certain conditions.

TL;DR: In this paper , a direct and model-free Gibbs posterior distribution for multivariate quantiles is proposed, which enjoys a root-n convergence rate and a Bernstein-von Mises property, i.e., for large n, the Gibbs posterior can be approximated by a Gaussian.

TL;DR: In this paper, the relationship and differences between stratification, rerandomization, and the combination of the two have not been previously investigated, and it is shown that stratified designs can be recreated by rerandomisation and explain why, in most cases, stratification on binary covariates followed by re-randomization on continuous covariates is more efficient than rerandomizing on all covariates at the same time.

TL;DR: In this paper , the authors proposed Bayesian D-optimal and Bayesian G-optimality models for the pairwise order model, which may have fewer runs than the number of parameters.

TL;DR: In this article , a series of optimal designs with flexible run sizes is proposed for designing order-of-addition experiments, where the full design can be partitioned into several isotopic Latin squares, and a fractional design is obtained by juxtaposing some selected Latin squares according to a particular optimality criterion.

TL;DR: In this paper, the authors consider design criteria which depend on several designs simultaneously and formulate equivalence theorems based on information matrices (if criteria depend on designs via information matrix) or with respect to the designs themselves (for finite design regions).

TL;DR: An asymptotic estimate for the p -value of an optimal test is found in the case where the alternative hypothesis is a known stationary ergodic source, and a family of tests each of which has the same asymPTotic estimate of the p-value for any (unknown) stationary er godic source is described.

TL;DR: In this article, a direct and model-free Gibbs posterior distribution for multivariate quantiles is proposed, which enjoys a root-n convergence rate and a Bernstein-von Mises property, i.e., for large n, the Gibbs posterior can be approximated by a Gaussian.

TL;DR: In this article, the authors established asymptotic results for the maximum likelihood and restricted maximum likelihood estimators of the parameters in the nested error regression model for clustered data when both of the number of independent clusters and the cluster sizes (the number of observations in each cluster) go to infinity.

TL;DR: Strong posterior consistency is shown to hold under notably weak regularity assumptions and adaptative convergence rates are obtained in terms of the approximation properties of positive linear operators generating the authors' models.

TL;DR: In this article , a vector autoregressive (VAR) model subject to Markov switching was considered and the fit of such a model and its capability to parametrize appropriately the covariance structure of the observed multivariate process was evaluated.

TL;DR: In this article , the temporal aggregation and systematic sampling (SS) for generalized autoregressive conditional heteroskedasticity (INGARCH) processes are discussed. But the authors focus on the weakness of the weak properties of INGARCH processes in terms of linear projection.

TL;DR: In this article, the authors proposed an estimator for the structural break location of high-dimensional time series, whose convergence rate is shown to depend on an interplay between the dimension of the observed time series and the strength of the underlying factor structure.

TL;DR: In this article , a monotone frequentist measure of evidence is proposed for testing variance components in a linear mixed model with a general parameter space that includes null variance components for the random effects, and the test based on the s-value has significance level α , i.e., the maximum probability of the s -value not exceeding a fixed threshold α does not surpass α for each fixed α ∈ ( 0 , 1 ) .

TL;DR: In this article , the relationship and differences between stratification, rerandomization, and the combination of the two have not been previously investigated, and it is shown that stratified designs can be recreated by rerandomisation and explain why, in most cases, stratification on binary covariates followed by re-randomization on continuous covariates is more efficient than rerandomizing on all covariates at the same time.

TL;DR: In this paper, the J-fold cross-validation criterion is developed to determine averaging weights, and the resulting estimators are asymptotically optimal under some regularity conditions.

TL;DR: In this article, the authors consider the estimation of two cumulative incidence functions, F 1 and F 2, corresponding to two competing risks when the ratio R ( t ) ≡ F 1 (t ) / F 2 (t) is nondecreasing in t > 0.

TL;DR: In this paper , a fully semiparametric Bayesian framework for DR causal inference was developed by bridging a nonparametric Bayes procedure with empirical likelihood theory via semi-parametric linear regression, which allows the posterior distribution of the causal parameter to be simulated via Markov chain Monte Carlo methods.

TL;DR: In this paper , the authors propose to incorporate spatial information by assigning intrinsic autoregressive priors to the logit prior probabilities of inclusion, which results in more similar shrinkage penalties among spatially adjacent parameters.

TL;DR: In this article, a conjecture optimal design for a k-level accelerated degradation test (ADT) design problem is proposed and the general equivalence theorem is applied to show that this conjecture design turns out to be the global V-optimal design.

TL;DR: Using MCMC to fit Bayesian models can be computationally prohibitive for large-scale data, but the model is fitted by adapting a computationally efficient coordinate-descent-based EM algorithm to incorporate spatial information by assigning intrinsic autoregressive priors to the logit prior probabilities of inclusion.

TL;DR: In this paper , the authors consider design criteria which depend on several designs simultaneously and formulate equivalence theorems based on information matrices (if criteria depend on designs via information matrix) or with respect to the designs themselves (for finite design regions).

TL;DR: In this paper , the authors derived Poisson limit theorems for the multinomial Cressie-Read goodness-of-fit statistics and some of their modifications under the assumption that as the sample size n goes to infinity, the number of groups mn increases with sample size.

TL;DR: In this article , a new class of models with copula is proposed to accurately and flexibly capture the correlation structure between two random coefficients in the binomial AR(1) process, and a real data example is provided to illustrate the model.