Search

We consider bias reduced estimators for the tail index and tail probabilities under random right censoring in case of Pareto-type distributions. The solution is based on second-order refined peaks-over-threshold modelling as developed in Beirlant et al. (2009). [ABSTRACT FROM AUTHOR]

Abstract: The aim of this paper is to study the consistency of the kernel density estimator pertaining to a continuous time stationary process , with an underlying density . More precisely, in a rather general dependency setting, where we use a martingale difference device and a technique based on a sequence of projections on -fields, we establish the almost sure pointwise and uniform consistencies with rates of the estimate of built upon the part of the process . [Copyright &y& Elsevier]

Abstract: This paper applied the maximization by parts (MBP) and the modified MBP (MMBP) methods to estimate the C-MGARCH model and compare the effectiveness of two methods. Monte Carlo simulation studies show that both MBP and MMBP methods are more efficient than that of the IFM method. [Copyright &y& Elsevier]

Abstract: In this paper, we propose a flexible model for cyclic abundance data, that takes into account predator–prey interactions. The idea is to incorporate a predator–prey dynamical model into a non homogeneous Poisson process of periodic intensity rate. Further, we devise an easy to implement estimation method and derive asymptotic confidence intervals. Finally, a simulation study indicates that the method performs well for small samples which are typical in ecology. [Copyright &y& Elsevier]

Abstract: In this paper, we investigate the estimation of conditional density function based on the single-index model for functional time series data. The asymptotic normality of the conditional density estimator and the conditional mode estimator for the mixing dependence functional time series data are obtained, respectively. Furthermore, as applications, the asymptotic confidence interval of the conditional density function and the conditional mode are also presented for . [Copyright &y& Elsevier]

Abstract: In this paper we consider the estimation problem for the quasi-likelihood model in presence of non-sample information (NSI). More specifically, we introduce a shrinkage estimation strategy for simultaneous model selection and parameter estimation by using the maximum quasi-likelihood estimates as the benchmark estimator, and define the pretest estimator (PTE), shrinkage estimator (SE) and positive-rule shrinkage estimator (PSE). Furthermore, we apply the lasso-type estimation strategy and compare the relative performance of lasso with the suggested estimators. The shrinkage estimators are shown to be efficient estimators compared to others. When the NSI is true the PTE has less risk compared to shrinkage and lasso estimators. [Copyright &y& Elsevier]

Abstract: With sparse structures and conditional independence, one could estimate the precision matrix of Gaussian graphical models more efficiently. studied objective priors for star-shape graphical models. We consider a generative star-shape model. Objective priors such as invariance priors, the Jeffreys prior and reference priors are derived for the proposed model. Closed form expressions of posterior distributions are derived based on a family of priors including the studied objective priors. [Copyright &y& Elsevier]

Abstract: We establish the law of the iterated logarithm for the product limit estimator, when the data are subject to double censoring. This investigation extends the results available for the model for singly censored data. [Copyright &y& Elsevier]

Abstract: An adjusted -squared type measure for exponential dispersion models, based on Kullback–Leibler divergences, is described and it is proved that the corrected measure is an asymptotically unbiased estimator of the population value, under the null hypotheses of no covariates. [Copyright &y& Elsevier]

Abstract: Suppose we have data from a bivariate model with parametric marginals. Efficient estimators of the parameters in the marginal models are generally not efficient in the bivariate model. In this article, we propose a method of improving these marginal estimators and demonstrate that the magnitude of this improvement can be as large as 100 percent in some cases. [Copyright &y& Elsevier]

Abstract: An alternative to the classical mixed model with normal random effects is to use a Dirichlet process to model the random effects. Such models have proven useful in practice, and we have observed a noticeable variance reduction, in the estimation of the fixed effects, when the Dirichlet process is used instead of the normal. In this paper we formalize this notion, and give a theoretical justification for the expected variance reduction. We show that for almost all data vectors, the posterior variance from the Dirichlet random effects model is smaller than that from the normal random effects model. [Copyright &y& Elsevier]

Abstract: An alternative criterion is presented for a linear function of the unknown parameters, in a linear model, to be estimable. If there is a linear relationship between members of a subset of the columns of the design matrix , then the parameters associated with all the columns in that subset are not estimable. Also any part of their linear relationship is non-estimable. [Copyright &y& Elsevier]