Biometrika

最新文章

• Information-theoretic optimality of observation-driven time series models for continuous responses AI翻译
  作者：Blasques, F.; Koopman, S. J.; Lucas, A.
  作者单位：Vrije Universiteit Amsterdam
  摘要：We investigate information-theoretic optimality properties of the score function of the predictive likelihood as a device for updating a real-valued time-varying parameter in a univariate observation-driven model with continuous responses. We restrict our attention to models with updates of one lag order. The results provide theoretical justification for a class of score-driven models which includes the generalized autoregressive conditional heteroskedasticity model as a special case. Our main...
• Singular value shrinkage priors for Bayesian prediction AI翻译
  作者：Matsuda, Takeru; Komaki, Fumiyasu
  作者单位：University of Tokyo
  摘要：We develop singular value shrinkage priors for the mean matrix parameters in the matrix-variate normal model with known covariance matrices. Our priors are superharmonic and put more weight on matrices with smaller singular values. They are a natural generalization of the Stein prior. Bayes estimators and Bayesian predictive densities based on our priors are minimax and dominate those based on the uniform prior in finite samples. In particular, our priors work well when the true value of the p...
• Optimal two-level choice designs for any number of choice sets AI翻译
  作者：Singh, Rakhi; Chai, Feng-Shun; Das, Ashish
  作者单位：Indian Institute of Technology System (IIT System); Indian Institute of Technology (IIT) - Bombay; Academia Sinica - Taiwan; Indian Institute of Technology System (IIT System); Indian Institute of Technology (IIT) - Bombay
  摘要：For two-level choice experiments, we obtain a simple form of the information matrix of a choice design for estimating the main effects, and provide D- and MS-optimal paired choice designs with distinct choice sets under the main effects model for any number of choice sets. It is shown that the optimal designs under the main effects model are also optimal under the broader main effects model. We find that optimal choice designs with a choice set size two often outperform their counterparts with...
• On the dependence structure of bivariate recurrent event processes: inference and estimation AI翻译
  作者：Ning, Jing; Chen, Yong; Cai, Chunyan; Huang, Xuelin; Wang, Mei-Cheng
  作者单位：University of Texas System; UTMD Anderson Cancer Center; University of Texas System; University of Texas Health Science Center Houston; University of Texas System; University of Texas Health Science Center Houston; Johns Hopkins University
  摘要：Bivariate or multivariate recurrent event processes are often encountered in longitudinal studies in which more than one type of event is of interest. There has been much research on regression analysis for such data, but little has been done to measure the dependence between recurrent event processes. We propose a time-dependent measure, termed the rate ratio, to assess the local dependence between two types of recurrent event processes. We model the rate ratio as a parametric function of tim...
• Big data and precision AI翻译
  作者：Cox, D. R.
  作者单位：University of Oxford
  摘要：So-called big data are likely to have complex structure, in particular implying that estimates of precision obtained by applying standard statistical procedures are likely to be misleading, even if the point estimates of parameters themselves may be reasonably satisfactory. While this possibility is best explored in the context of each special case, here we outline a fairly general representation of the accretion of error in large systems and explore the possible implications for the estimatio...
• Efficient inference and simulation for elliptical Pareto processes AI翻译
  作者：Thibaud, Emeric; Opitz, Thomas
  作者单位：Colorado State University System; Colorado State University Fort Collins; INRAE
  摘要：Recent advances in extreme value theory have established l-Pareto processes as the natural limits for extreme events defined in terms of exceedances of a risk functional. In this paper we provide methods for the practical modelling of data based on a tractable yet flexible dependence model. We introduce the class of elliptical l-Pareto processes, which arise as the limits of threshold exceedances of certain elliptical processes characterized by a correlation function and a shape parameter. An ...
• Efficient estimation of nonparametric genetic risk function with censored data AI翻译
  作者：Wang, Yuanjia; Liang, Baosheng; Tong, Xingwei; Marder, Karen; Bressman, Susan; Orr-Urtreger, Avi; Giladi, Nir; Zeng, Donglin
  作者单位：Beijing Normal University; Columbia University; Harvard University; Harvard University Medical Affiliates; Beth Israel Deaconess Medical Center; Tel Aviv University; Sackler Faculty of Medicine; University of North Carolina; University of North Carolina Chapel Hill
  摘要：With the discovery of an increasing number of causal genes for complex human disorders, it is crucial to assess the genetic risk of disease onset for individuals who are carriers of these causal mutations and to compare the distribution of the age-at-onset for such individuals with the distribution for noncarriers. In many genetic epidemiological studies that aim to estimate causal gene effect on disease, the age-at-onset of disease is subject to censoring. In addition, the mutation carrier or...
• Space-filling properties of good lattice point sets AI翻译
  作者：Zhou, Yongdao; Xu, Hongquan
  作者单位：Sichuan University; University of California System; University of California Los Angeles
  摘要：We study space-filling properties of good lattice point sets and obtain some general theoretical results. We show that linear level permutation does not decrease the minimum distance for good lattice point sets, and we identify several classes of such sets with large minimum distance. Based on good lattice point sets, some maximin distance designs are also constructed.
• Moment-type estimators for the proportional likelihood ratio model with longitudinal data AI翻译
  作者：Luo, Xiaodong; Tsai, Wei Yann
  作者单位：Icahn School of Medicine at Mount Sinai; Columbia University
  摘要：Luo & Tsai, Biometrika 99, 211-22, 2012, proposed a proportional likelihood ratio model and discussed a maximum likelihood method for its parameter estimation. In this paper, we use this model as the marginal distribution to analyse longitudinal data, where the maximum likelihood method is not directly applicable because the joint distribution is not fully specified. We propose a moment-type method that is an extension of the generalized estimating equation method. The resulting estimators are...
• On the degrees of freedom of reduced-rank estimators in multivariate regression AI翻译
  作者：Mukherjee, A.; Chen, K.; Wang, N.; Zhu, J.
  作者单位：University of Connecticut; University of Michigan System; University of Michigan
  摘要：We study the effective degrees of freedom of a general class of reduced-rank estimators for multivariate regression in the framework of Stein's unbiased risk estimation. A finite-sample exact unbiased estimator is derived that admits a closed-form expression in terms of the thresholded singular values of the least-squares solution and hence is readily computable. The results continue to hold in the high-dimensional setting where both the predictor and the response dimensions may be larger than...