检索结果-内蒙古大学图书馆

A MCMC algorithm to fit a general exchangeable model

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION 1996年第3期25卷 573-592页

作者： Albert, J BOWLING GREEN STATE UNIV DEPT MATH & STATBOWLING GREENOH 43403

Consider the exchangeable Bayesian hierarchical model where observations y(i) are independently distributed from sampling densities with unknown means, the means mu(i) are a random sample from a distribution g, and the parameters of g are assigned a known distribution h. A simple algorithm is presented for summarizing the posterior distribution based on Gibbs sampling and the metropolis algorithm. The software program Matlab is used to implement the algorithm and provide a graphical output analysis. An binomial example is used to illustrate the flexibility of modeling possible using this algorithm. Methods of model checking and extensions to hierarchical regression modeling are discussed.

关键词： hierarchical model Gibbs sampler metropolis algorithm logistic model output analysis

来源：评论

学校读者我要写书评

暂无评论

Bayesian inference for random coefficient dynamic panel data models

引用

JOURNAL OF APPLIED STATISTICS 2017年第9期44卷 1543-1559页

作者： Liu, Fang Zhang, Peng Erkan, Ibrahim Small, Dylan S. Merck & Co Inc Upper Gywnedd PA USA XHTH Asset Management Beijing Peoples R China Ankara Dev Agcy Ankara Turkey Univ Penn Dept Stat Philadelphia PA 19104 USA

We develop a hierarchical Bayesian approach for inference in random coefficient dynamic panel data models. Our approach allows for the initial values of each unit's process to be correlated with the unit-specific coefficients. We impose a stationarity assumption for each unit's process by assuming that the unit-specific autoregressive coefficient is drawn from a logitnormal distribution. Our method is shown to have favorable properties compared to the mean group estimator in a Monte Carlo study. We apply our approach to analyze energy and protein intakes among individuals from the Philippines.

关键词： Dynamic panel data Bayesian inference Gibbs sampling metropolis algorithm

来源：评论

学校读者我要写书评

暂无评论

A hybrid system of heuristics and simulated annealing for large size scheduling problems of batch processes with pipeless intermediate storage

引用

KOREAN JOURNAL OF CHEMICAL ENGINEERING 1997年第4期14卷 225-232页

作者： Kim, HJ Jung, JH Kim, M Lee, IB YEUNGNAM UNIV SCH CHEM ENGN & TECHNOLKYONGSAN 712749SOUTH KOREA

This paper describes a new scheduling solution for large number multi-product batch processes with complex intermediate storage system. Recently many batch chemical industries have turned their attention to a more efficient system known as a pipeless batch system. But existing plants need to change their systems to pipeless systems, piece by piece. In this case, current systems are changed to pipeless systems by way of non critical process operations such as through the use of intermediate storage. We have taken the conventional batch plant with a pipeless storage system as an objective process. Although the operation of a pipeless storage system becomes more complex, its efficiency is very high. With this system, all of the storage should be commonly used by any batch unit. For this reason, solving the optimal scheduling problem of this system with a mathematical method is very difficult. Despite the attempts of many previous researches, there has been no contribution which solves the scheduling of intermediate storage for complex batch processes. In this paper, we have developed a hybrid system of heuristics and Simulated Annealing (SA) for large multi-product processes using a pipeless storage system. The results of this study show that the performance and computational time of this method are superior to that of SA and Rapid Access Extensive Search (RAES) methods.

关键词： a hybrid system of heuristics and simulated annealing pipeless intermediate storage metropolis algorithm probability of up-hill movement

来源：评论

学校读者我要写书评

暂无评论

A theory for dynamic weighting in Monte Carlo computation

引用

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION 2001年第454期96卷 561-573页

作者： Liu, JS Liang, FM Wong, WH Harvard Univ Dept Stat Cambridge MA 02138 USA Natl Univ Singapore Dept Stat & Appl Probabil Singapore 117548 Singapore Harvard Univ Dept Biostat & Stat Cambridge MA 02138 USA

This article provides a first theoretical analysis of a new Monte Carlo approach, the dynamic weighting algorithm, proposed recently by Wong and Liang. In dynamic weighting Monte Carlo, one augments the original stale space of interest by a weighting factor, which allows the resulting Markov chain to move more freely and to escape from local modes. II uses a new invariance principle to guide the construction of transition rules. We analyze the behavior of the weights resulting from such a process and provide detailed recommendations on how to use these weights properly. Our recommendations;are supported by a renewal theory-type analysis. Our theoretical investigations are further demonstrated by a simulation study and applications in neural network training and Ising model simulations.

关键词： Gibbs sampling importance sampling Ising model metropolis algorithm neural network renewal theory simulated annealing simulated tempering

来源：评论

学校读者我要写书评

暂无评论

Estimating incumbency advantage and its variation, as an example of a before-after study

引用

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION 2008年第482期103卷 437-446页

作者： Gelman, Andrew Huang, Zaiying Columbia Univ Dept Stat New York NY 10027 USA Columbia Univ Dept Polit Sci New York NY 10027 USA New York Times Co Circulat Dept New York NY USA

Incumbency advantage is one of the most widely studied features in American legislative elections. In this article we construct and implement an estimate that allows incumbency advantage to vary between individual incumbents. This model predicts that open-seat elections will be less variable than those with incumbents running, an observed empirical pattern that is not explained by previous models. We apply our method to the U.S. House of Representatives in the twentieth century. Our estimate of the overall pattern of incumbency advantage over time is similar to previous estimates (although slightly lower), and we also find a pattern of increasing variation. More generally, our multilevel model represents a new method for estimating effects in before-after studies.

关键词： Bayesian inference before-after study congressional election Gibbs sampler incumbency advantage metropolis algorithm multilevel model

来源：评论

学校读者我要写书评

暂无评论

On the geometrical convergence of Gibbs sampler in R^d

引用

JOURNAL OF MULTIVARIATE ANALYSIS 1998年第1期66卷 22-37页

作者： Hwang, CR Sheu, SJ Acad Sinica Math Inst Taipei Taiwan

The geometrical convergence of the Gibbs sampler for simulating a probability distribution in R-d is proved. The distribution has a density which is a bounded perturbation of a log-concave function and satisfies some growth conditions. The analysis is based on a representation of the Gibbs sampler and some powerful results from the theory of Harris recurrent Markov chains. (C) 1998 Academic Press.

关键词： stochastic relaxation Gibbs sampler Markov chain geometrical convergence Harris recurrence Monte Carlo Markov chain metropolis algorithm nonlinear autoregression

来源：评论

学校读者我要写书评

暂无评论

Development of serum parameters panels for the early detection of pancreatic cancer

引用

INTERNATIONAL JOURNAL OF CANCER 2014年第11期134卷 2646-2655页

作者： Zhang, Pengjun Zou, Meng Wen, Xinyu Gu, Feng Li, Juan Liu, Gaixia Dong, Jingxiao Deng, Xinxin Gao, Jing Li, Xiaolong Jia, Xingwang Dong, Zhennan Chen, Luonan Wang, Yong Tian, Yaping Chinese Peoples Liberat Army Gen Hosp Dept Clin Biochem State Key Lab Kidney Dis Beijing Peoples R China Chinese Acad Sci Acad Math & Syst Sci Beijing Peoples R China Nankai Univ Coll Med Tianjin 300071 Peoples R China Chinese Peoples Liberat Army Gen Hosp Dept Radiol Beijing Peoples R China Chinese Acad Sci Natl Ctr Math & Interdisciplinary Sci Beijing Peoples R China Chinese Acad Sci Shanghai Inst Biol Sci SIBS Novo Nordisk Translat Res Ctr PreDiabet Key Lab Syst Biol Shanghai Peoples R China

Early detection of pancreatic cancer is promising for improving clinical outcome;however, no effective biomarker has yet been identified. Here, we detected 61 clinical serum parameters in 200 healthy controls (Ctrls), 163 pancreatic ductal adenocarcinoma (PDAC) patients and 109 benign pancreatitis patients (Benign) in the training group. A metropolis algorithm with Monte Carlo simulation was used for identifying parameter panels. Sera from 183 Ctrl, 129 PDAC and 95 Benign individuals were used for cross-validation. Samples from 77 breast, 72 cervical, 101 colorectal, 138 gastric, 108 prostate and 132 lung cancer patients were collected for evaluating cancer selectivity. A panel consisting of carbohydrate antigen (CA)19-9, albumin (ALB), C-reactive protein (CRP) and interleukin (IL)-8 had the highest diagnostic value for discriminating between PDAC and Ctrl. The sensitivity (SN) was 99.39% for all-stage, 96.10% for early-stage and 98.80% for advanced-stage PDAC at 90% specificity (SP). In the validation group, the sensitivities were 93.80, 93.10 and 94.40%, respectively, at 90% SP. This panel also identified 80.52% of the breast cancer, 66.67% cervical cancer, 86.14% colorectal cancer, 89.86% gastric cancer, 71.30% prostate cancer and 93.85% lung cancer samples as non-PDAC. The panel consisting of CA19-9, carbon dioxide, CRP and IL-6 panel had the highest diagnostic value for discriminating between PDAC and Benign. The SN was 74.23% for all-stage, 75.30% for early-stage and 74.40% for advanced-stage PDAC at 90% SP. In the validation group, the sensitivities were 72.10, 76.10 and 67.20%, respectively, at 90% SP. Our parameter panels may aid in the early detection of PDAC to improve clinical outcome. What's new? Few people survive a pancreatic cancer diagnosis, in part because it's rarely detected early. In this study, the authors went looking for biomarkers that could indicate a developing cancer. They tested 61 different biomarkers in blood samples from patients with

关键词： serum pancreatic cancer metropolis algorithm Monte Carlo simulation early detection

来源：评论

学校读者我要写书评

暂无评论

On adaptive Markov chain Monte Carlo algorithms

引用

BERNOULLI 2005年第5期11卷 815-828页

作者： Atchadé, YF Rosenthal, JS Univ Ottawa Dept Math & Stat Ottawa ON K1N 6NS Canada Univ Toronto Dept Stat Toronto ON M5S 3G3 Canada

We look at adaptive Markov chain Monte Carlo algorithms that generate stochastic processes based on sequences of transition kernels, where each transition kernel is allowed to depend on the history of the process. We show under certain conditions that the stochastic process generated is ergodic, with appropriate stationary distribution. We use this result to analyse an adaptive version of the random walk metropolis algorithm where the scale parameter sigma is sequentially adapted using a Robbins-Monro type algorithm in order to find the optimal scale parameter sigma(opt). We close with a simulation example.

关键词： adaptive Markov chain Monte Carlo metropolis algorithm mixingales parameter tuning Robbins-Monro algorithm

来源：评论

学校读者我要写书评

暂无评论

Comparison of two methods for calculating the partition functions of various spatial statistical models

引用

AUSTRALIAN & NEW ZEALAND JOURNAL OF STATISTICS 2001年第1期43卷 47-65页

作者： Huang, F Ogata, Y Deakin Univ Sch Comp & Math Clayton Vic 3168 Australia Deakin Univ Sch Comp & Math Tokyo 106 Japan Inst Stat Math Tokyo 106 Japan

Likelihood computation in spatial statistics requires accurate and efficient calculation of the normalizing constant (i.e. partition function) of the Gibbs distribution of the model. Two available methods to calculate the normalizing constant by Markov chain Monte Carlo methods are compared by simulation experiments for an Ising model, a Gaussian Markov field model and a pairwise interaction point field model.

关键词： Gibbs sampling likelihood MCMC integration metropolis algorithm partition function

来源：评论

学校读者我要写书评

暂无评论

Dose response modelling of Escherichia coli O157 incorporating data from foodborne and environmental outbreaks

引用

INTERNATIONAL JOURNAL OF FOOD MICROBIOLOGY 2005年第1期103卷 35-47页

作者： Strachan, NJC Doyle, MP Kasuga, F Rotariu, O Ogden, ID Univ Aberdeen Sch Biol Sci Aberdeen AB24 3UU Scotland Univ Georgia Ctr Food Safety Griffin GA USA Natl Inst Hlth Sci Div Biomed Food Res Setagaya Ku Tokyo 1588501 Japan Univ Aberdeen Sch Med Dept Med Microbiol Appl Food Microbiol Grp Aberdeen AB25 2ZD Scotland

A human dose response model for Escherichia coli O157 would enable prediction of risk of infection to humans following exposure from either foodborne or environmental pathways. However, due to the severe nature of the disease, volunteer human dose response studies cannot be carried out. Surrogate models from Shigella fed to humans and E. coli O157 to rabbits have been utilised but are significantly different to one another. In addition data obtained by animal exposure may not be representative for human beings. An alternative approach to generating and validating a dose response model is to use quantitative data obtained from actual human outbreaks. This work collates outbreak data obtained from global sources and these are fitted using exponential and beta-Poisson models. The best fitting model was found to be the beta-Poisson model using a beta-binomial likelihood and the authors favour the exact version of this model. The confidence levels in this model encompass a previously published Shigella dose response model. The potential incorporation of this model into QMRAs is discussed together with applications of the model to help explain foodborne outbreaks. (c) 2005 Elsevier B.V. All rights reserved.

关键词： dose response E. coli O157 food outbreaks environmental outbreaks risk assessment Markov Chain Monte Carlo metropolis algorithm epidemiology

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：