检索结果-内蒙古大学图书馆

Too many covariates and too few cases? - a comparative study

STATISTICS IN MEDICINE 2016年第25期35卷 4546-4558页

作者： Chen, Qingxia Nian, Hui Zhu, Yuwei Talbot, H. Keipp Griffin, Marie R. Harrell, Frank E., Jr. Vanderbilt Univ Sch Med Dept Biostat Nashville TN 37232 USA Vanderbilt Univ Sch Med Dept Biomed Informat Nashville TN 37232 USA Vanderbilt Univ Sch Med Dept Med Nashville TN 37232 USA Vanderbilt Univ Sch Med Dept Hlth Policy Nashville TN 37232 USA VA TN Valley Hlth Care Syst Med Midsouth Geriatr Res Educ & Clin Ctr Nashville TN 37232 USA VA TN Valley Hlth Care Syst Clin Res Ctr Excellence Nashville TN 37232 USA

Prior research indicates that 10-15 cases or controls, whichever fewer, are required per parameter to reliably estimate regression coefficients in multivariable logistic regression models. This condition may be difficult tomeet even in a well-designed study when the number of potential confounders is large, the outcome is rare, and/or interactions are of interest. Various propensity score approaches have been implemented when the exposure is binary. Recent work on shrinkage approaches like lasso were motivated by the critical need to develop methods for the p >> n situation, where p is the number of parameters and n is the sample size. Those methods, however, have been less frequently used when p approximate to n, and in this situation, there is no guidance on choosing among regular logistic regression models, propensity score methods, and shrinkage approaches. To fill this gap, we conducted extensive simulations mimicking our motivating clinical data, estimating vaccine effectiveness for preventing influenza hospitalizations in the 2011-2012 influenza season. Ridge regression and penalized logistic regression models that penalize all but the coefficient of the exposure may be considered in these types of studies. Copyright (C) 2016 John Wiley & Sons, Ltd.

关键词： lasso logistic regression model over-parameterization propensity score ridge

来源：评论

学校读者我要写书评

暂无评论

On the repeated measures designs and sample sizes for randomized controlled trials

引用

BIOSTATISTICS 2016年第2期17卷 334-349页

作者： Tango, Toshiro Ctr Med Stat Minato Ku 2-9-6 Higashi Shimbashi Tokyo 1050021 Japan

For the analysis of longitudinal or repeated measures data, generalized linear mixed-effects models provide a flexible and powerful tool to deal with heterogeneity among subject response profiles. However, the typical statistical design adopted in usual randomized controlled trials is an analysis of covariance type analysis using a pre-defined pair of "pre-post" data, in which pre-(baseline) data are used as a covariate for adjustment together with other covariates. Then, the major design issue is to calculate the sample size or the number of subjects allocated to each treatment group. In this paper, we propose a new repeated measures design and sample size calculations combined with generalized linear mixed-effects models that depend not only on the number of subjects but on the number of repeated measures before and after randomization per subject used for the analysis. The main advantages of the proposed design combined with the generalized linear mixed-effects models are (1) it can easily handle missing data by applying the likelihood-based ignorable analyses under the missing at random assumption and (2) it may lead to a reduction in sample size, compared with the simple pre-post design. The proposed designs and the sample size calculations are illustrated with real data arising from randomized controlled trials.

关键词： Likelihood-based ignorable analysis logistic regression model Missing data Poisson regression model Pre-post design

来源：评论

学校读者我要写书评

暂无评论

Maximum prey size estimation of longtooth grouper, Epinephelus bruneus, using morphological features, and predation experiments on juvenile cannibalism

引用

AQUACULTURE RESEARCH 2016年第2期47卷 605-611页

作者： Inoue, Nariaki Satoh, Jun Mekata, Tohru Iwasaki, Takashi Mori, Koichiro Fisheries Res Agcy Natl Res Inst Aquaculture Kamiura Lab Saiki Oita 8722602 Japan Fisheries Res Agcy Seikai Natl Fisheries Res Inst Yaeyama Lab Ishigaki Okinawa Japan

In fish culture, cannibalism, which induces critical damage, is separated into two types: type I, which occurs in early larval stages when size heterogeneity is low;and type II, which starts in later larval and/or early juvenile stages following type I when size heterogeneity is high. In type II, the cannibal's maximum prey size is limited by its gape size relative to the prey fish body size. Ratios of prey to cannibal body size changes are associated with allometric factors and feeding behaviour with growth. In a previous study on grouper, the maximal prey size was estimated morphologically (i.e., the relationship between gape size and body height). This method is very simple and useful;however, it does not consider allometric factors and feeding behaviour changes;therefore, it probably causes a large estimation error. To overcome this, we present a new method for estimating maximal prey size to determine the size ranges for size grading of longtooth grouper, Epinephelus bruneus, juveniles using predation experiments. It is shown that grouper type II cannibalism is affected by two factors;in early juvenile stages by morphology and in later (TL over 50mm) by prey size selectivity.

关键词： Cannibalism Epinephelus bruneus mouth size logistic regression model aquaculture

来源：评论

学校读者我要写书评

暂无评论

A rapid method for flood susceptibility mapping in two districts of Phatthalung Province (Thailand): present and projected conditions for 2050

引用

NATURAL HAZARDS 2016年第1期81卷 329-346页

作者： Marconi, Michele Gatto, Beatrice Magni, Michele Marincioni, Fausto Univ Politecn Marche Dept Life & Environm Sci Via Brecce Bianche I-60131 Ancona Italy

This study discusses the application of a multiple logistic regression analysis in Khao Chai Son and Mueang Phatthalung districts (Phatthalung Province in southern Thailand), which were the two worst flooded districts in the 2011 inundation. The aim is to test an easy, rapid, and cost-effective method to asses flood susceptibility in a data-poor country. Climatic, topographic, and geological data have been overlaid with those of the flood events occurred in the study area from 2007 to 2011. Results showed a positive spatial correlation between the northeast monsoon precipitation and flooding. Moreover, using the rainfall projection of the U.S. National Center for Atmospheric Research the proposed model forecasts a sharp increase of flood susceptibility in the study area by the year 2050. Given the versatility of such model, local governments could easily use it to define the areas in their territories most exposed to flood hazard and timely implement risk reduction policies and practices.

关键词： Climate change Disaster risk reduction Flood susceptibility logistic regression model Thailand

来源：评论

学校读者我要写书评

暂无评论

Factors determining child mortality in Uganda

引用

INTERNATIONAL JOURNAL OF SOCIAL ECONOMICS 2017年第5期44卷 633-642页

作者： Kaberuka, Will Mugarura, Alex Tindyebwa, Javan Bishop, Debra S. Makerere Univ Business Sch Dept Management Sci Kampala Uganda Makerere Univ Business Sch Dept Accounting & Finance Kampala Uganda Drake Univ Coll Business & Publ Adm Des Moines IA 50311 USA

Purpose - The purpose of this paper is to establish socio-economic factors and maternal practices that determine child mortality in Uganda. Design/methodology/approach - The paper examines the role of sex, birth weight, birth order and duration of breastfeeding of a child;age, marital status and education of the mother;and household wealth in determining child mortality. The study employs a logistic regression model to establish which of the factors significantly impacts child mortality in Uganda. Findings - The study established that education level, age and marital status of the mother as well as household wealth significantly impact child mortality. Also important are the sex, birth weight, birth order and breastfeeding duration. Research limitations/implications - Policies aimed at promoting breastfeeding and education of female children can make a significant contribution to the reduction of child mortality in Uganda. Practical implications - Health care intervention programs should focus on single, poor and uneducated mothers as their children are at great risk due to poor and inadequate health care utilization. Originality/value - This paper could be the first effort in examining child mortality status in Uganda using a logistic regression model.

关键词： Child mortality Exclusive breastfeeding logistic regression model Under-five mortality rate

来源：评论

学校读者我要写书评

暂无评论

The Analysis of Performance of Total Amount Reduction of Pollutants Emission Based on logistic regression model

The Analysis of Performance of Total Amount Reduction of Pol...

引用

Biennial International Conference on Ecological Informatics and Ecosystem Conservation (ISEIS)

作者： Wang Chunmei Sun Dezhi Wang Zhen Lin Zhaolan Beijing Forestry Univ Coll Environm Sci & Engn Beijing 100083 Peoples R China

Environmental policy is one of the main tools of environmental management for the government, it is very important for performance evaluation. This study is quantitative analysis for the relationship between the various types of environmental policies and the performance of total COD in wastewater emission volumes reducing, to obtain the role of pollution reduction from the support of policy. Based on the results of logistic model analysis, 5 reduction measures significantly affected the performance in three types of emission reduction targets. Among them, shutting down and relocate the number of enterprises have the greatest contribution to the emission reduction performance, and it followed by the sewage treatment rate, the removing of COD by the new wastewater treatment facilities, the number of projects completed under deadline management, the proportion of secondary industry. The frame measures are very effective and eliminating trailed productivity is the highest efficiency measure in all pollution control measures. Sewage treatment measures in this area also played a key role in pollution reduction;the sewage charging policy plays a key role in economic management. The adjustment of economic structure is concerned with pollution reduction. The more proportion of secondary industry covers, the more difficulty emission reducing is. Therefore, the government should pay more attention on the adjustment of industrial structure and strengthen the sewage treatment capacity to improve the performance of total amount reduction of pollutants emission.

关键词： logistic regression model Total Amount Reduction of Pollutants Emission Analysis of performance

来源：评论

学校读者我要写书评

暂无评论

Machine-Learning-Based Prediction of a Missed Scheduled Clinical Appointment by Patients with Diabetes

引用

Journal of Diabetes Science and Technology 2016年第3期10卷 730-736页

作者： Kurasawa, Hisashi Hayashi, Katsuyoshi Fujino, Akinori Takasugi, Koichi Haga, Tsuneyuki Waki, Kayo Noguchi, Takashi Ohe, Kazuhiko University of Tokyo 7-3-1 Hongo Bunkyoku Tokyo 113-8655 Japan Nippon Telegraph and Telephone Corporation Tokyo Japan

Background: About 10% of patients with diabetes discontinue treatment, resulting in the progression of diabetes-related complications and reduced quality of life. Objective: The objective was to predict a missed clinical appointment (MA), which can lead to discontinued treatment for diabetes patients. Methods: A machine-learning algorithm was used to build a logistic regression model for MA predictions, with L2-norm regularization used to avoid over-fitting and 10-fold cross validation used to evaluate prediction performance. Data associated with patient MAs were extracted from electronic medical records and classified into two groups: one related to patients' clinical condition (X1) and the other related to previous findings (X2). The records used were those of the University of Tokyo Hospital, and they included the history of 16 026 clinical appointments scheduled by 879 patients whose initial clinical visit had been made after January 1, 2004, who had diagnostic codes indicating diabetes, and whose HbA1c had been tested within 3 months after their initial visit. Records between April 1, 2011, and June 30, 2014, were inspected for a history of MAs. Results: The best predictor of MAs proved to be X1 + X2 (AUC = 0.958);precision and recall rates were, respectively, 0.757 and 0.659. Among all the appointment data, the day of the week when an appointment was made was most strongly associated with MA predictions (weight = 2.22). Conclusions: Our findings may provide information to help clinicians make timely interventions to avoid MAs. © Diabetes Technology Society.

关键词： L2-norm regularization logistic regression model machine learning missed clinic appointment

来源：评论

学校读者我要写书评

暂无评论

Analysis of Factors Contributing to Hit-and-Run Crashes Involved with Improper Driving Behaviors

Analysis of Factors Contributing to Hit-and-Run Crashes Invo...

引用

6th International Conference on Green Intelligent Transportation System and Safety (GITSS)

作者： Zhou, Bei Roshandeh, Arash M. Zhang, Shengrui Ma, Zhuanglin Changan Univ Shcool Highway South 2nd Ring Rd Xian 710064 Peoples R China City Alpharetta Traff Control Ctr 1790 Hembree Rd Alpharetta GA 30009 USA

Hit-and-run crashes are accidents where drivers of striking vehicles fail to stop after crashes. Without helping victims or reporting accidents to associated authorities could increase the likelihood of serious injuries and even fatalities. In order to reduce hit-and-run crashes, it is important to understand factors contributing to decisions of fleeing crash scenes. In current study, various factors which could affect occurrences of hit-and-run crashes were thoroughly investigated against six different improper driving behaviors. logistic regression models were established to facilitate the analysis. Police -reported crash data within Cook County, Illinois, USA between 2004 and 2012 were used in this study. The results showed that variables contributing to hit-and-run crashes varied for different improper driving behaviors. Among six established models, "following too closely" and "distraction by phone" models had most statistically significant variables. This study also concluded that following variables would increase the likelihood of hit-and-run crashes in at least one model: multiple vehicle crash, weekend, population of 2,500 - 5,000, population of 5,000 - 10,000, national highway system, traffic signal, yield sign, shoulder, darkness, and less than three lanes. The results of current study could offer important insights for reducing hit-and-run crashes in both planning and operational levels. (C) 2016 The Authors. Published by Elsevier Ltd.

关键词： hit-and-run logistic regression model odds ratio traffic safety improper driving behaviors

来源：评论

学校读者我要写书评

暂无评论

Using Structural Features to Characterize Social Ties 1

Using Structural Features to Characterize Social Ties

引用

1st IEEE International Conference on Data Science in Cyberspace (DSC)

作者： Zuo, Yang Zhang, Kan Peking Univ Sch Elect Engn & Comp Sci Beijing Peoples R China

ISBN: (纸本)9781509011926

Social ties are crucial to understand group behaviors in social networks. Since users rarely label their friends explicitly in social networks, characterizing social ties and understanding their strength of these ties is a critical problem. In this paper, we apply logistic regression model to infer the strength of social ties based on the structural features of social networks, and we produce two new features: cliqueness and linkness. In addition, we use the Strong Triadic Closure principle as a global constraint. We test our model on four different datasets: Facebook, American College Football, Les Miserables and Zachary's Karate Club. Experimental results demonstrate that the performance of our model with structural features is satisfactory.

关键词： social networks social ties structural features logistic regression model

来源：评论

学校读者我要写书评

暂无评论

Marketing of fresh produce by smallholder farmers: A case study of uThungulu District Municipality, KwaZulu-Natal, South Africa

Marketing of fresh produce by smallholder farmers: A case st...

引用

作者： Ntokozo Mdlalose University of KwaZulu-Natal

学位级别：硕士

Smallholder farmers in South Africa and in other developing countries often have limited access to factors of production and information. Poor infrastructure, lack of marketing knowledge and lack of transport are amongst the many market-related constraints that smallholder farmers are facing. Consequently, such constraints limit their ability to access markets. This study was conducted in uThungulu District Municipality based on a sample of 80 farmers who were selected using stratified random sampling. The study made use of a structured questionnaire to collect data through face-to-face interviews. A binary logistic regression model was used to identify factors influencing marketing participation decisions among the sampled farmers. The empirical results showed that extension services contact, quality of roads to market, access to transport, distance to market, and timing of seeking markets had a statistically significant influence on market participation decisions. These findings suggest that an improvement in each of the significant variables can significantly influence farmer market participation decisions. Furthermore, based on the empirical results of the study, policy recommendations are suggested. These include improving dissemination of market information to farmers, encouraging farmers to seek markets before production, and encouraging farmers to add value to their produce.

关键词： Smallholder farmers market access market participation decision logistic regression model uThungulu District Municipality

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：