东北大学：某学院应用统计学专业《高级统计建模》课程教学大纲

高级统计建模课程教学大纲Generalised Linear Models Subject Syllabus一、课程信息SubjectInformation课程编号：开课学期：53100313008Subject IDSemester课程分类：所属课群：专业教育PA专业平台MTSectionCategory课程学分：总学时/周：3.556Credit PointsTotal Hours/Weeks理论学时：实验学时：3224LECT. HoursEXP. HoursPBL学时：实践学时/周：00PBL HoursPRAC. Hours/Weeks开课学院：适用专业：东北大学应用统计学ASCollegeStream悉尼智能科技学院课程属性：课程模式：引进 UTS必修CompulsoryPatternMode课程协调人：李晓奇成绩记载方式：百分制MarksCoordinatorLiXiaoqiResult Type先修课程：概率论与随机变量ProbabilityTheoryandRandomVariablesRequisites应用回归分析RegressionAnalysis1.Dobson, A. J.. (2002)An Introduction to Generalized Linear Models英文参考教材：(Links to an external site.)Linksto an external site,2nd Edition,CRCEN Textbooks2. Draper, N.R., Smith, H. (1998) Applied Regression Analysis, 3rdedition, Wiley.中文参考教材：无CN Textbooks教学资源：https:/www.sas.com/en_au/software/on-demand-for-academics.htmlResources(Links to an external site.)课程负责人(撰写李晓奇提交日期：人)4/29/2023LiXiaoqiSubmitted DateSubject Director任课教师(含负责詹姆斯·布朗、李晓奇、张建波、韩鹏人):JamesBrown、LiXiaoqi、ZhangJianbo、HanPengTaught by审核人：批准人：韩鹏史闻博Checked byApprovedby批准日期：单击或点击此处输Approved Date入日期

高级统计建模 课程教学大纲 Generalised Linear Models Subject Syllabus 一、课程信息 Subject Information 课程编号: Subject ID 3100313008 开课学期: Semester 5 课程分类: Category 专业教育 PA 所属课群: Section 专业平台 MT 课程学分: Credit Points 3.5 总学时/周: Total Hours/Weeks 56 理论学时: LECT. Hours 32 实验学时: EXP. Hours 24 PBL 学时: PBL Hours 0 实践学时/周: PRAC. Hours/Weeks 0 开课学院: College 东北大学 悉尼智能科技学院 适用专业: Stream 应用统计学 AS 课程属性: Pattern 必修 Compulsory 课程模式: Mode 引进 UTS 课程协调人: Coordinator 李晓奇 Li Xiaoqi 成绩记载方式: Result Type 百分制 Marks 先修课程: Requisites 概率论与随机变量 Probability Theory and Random Variables 应用回归分析 Regression Analysis 英文参考教材: EN Textbooks 1. Dobson, A. J. (2002) An Introduction to Generalized Linear Models (Links to an external site.) Links to an external site., 2nd Edition, CRC. 2. Draper, N.R., Smith, H. (1998) Applied Regression Analysis, 3rd edition, Wiley. 中文参考教材: CN Textbooks 无 教学资源: Resources https://www.sas.com/en_au/software/on-demand-for-academics.html (Links to an external site.) 课程负责人(撰写 人): Subject Director 李晓奇 Li Xiaoqi 提交日期: Submitted Date 4/29/2023 任课教师(含负责 人): Taught by 詹姆斯·布朗、李晓奇、张建波、韩鹏 James Brown、Li Xiaoqi、Zhang Jianbo、Han Peng 审核人: Checked by 韩鹏 批准人: Approved by 史闻博 批准日期: Approved Date 单击或点击此处输 入日期

二、教学目标SubjectLearningObjectives(SLOs)注：毕业要求及指标点可参照悉尼学院本科生培养方案，可根据实际情况增减行数Note: GA and index can be referred from undergraduate program in SSTC website.Please add/reduce lines based on subject.高级统计建模是应用统计学专业的专业方向类课程，主要包括一元线性回归、多元线性回归、广义线性模型、逻辑回归、顺序逻辑回归、多项逻辑回归、泊松回归等。本课程培养学生运用回归和分类模型解决复杂问题的能力，对取得的成果进行评价和分析，并以各种方式向不同的受众（专家和非专家）简洁准确地整体目表达信息、推理和结论。标：Advanced Statistical Modeling is aprofessional direction course for Applied Statistics majors, which mainlyOverallincludes unitary linear regression, multiple linear regression, generalized linear model, logical regression, OrdinalObjectivLogistic Regression, Multinomial Logistic Regression, Poisson regression, ete. This course cultivates studentseability to use regression and classification models to solve complex problems, evaluate and analyze the resultsachieved, and express information, reasoning, and conclusions succinctly and accurately to different audiences(experts and nonexperts) in various ways,掌握各类回归模型的建模条件和方法，会对模型进行评价，并运用SAS在计算机实现。1-1Master the modeling conditions and methods of various regression models, be able to evaluate themodels, and use SAS to implement them on a computer.培养研究技能和解决问题的能力，在证据的基础上进行论证，并在选择方法的基础上进行模拟。(1) 专Cultivate research skills and problem-solving abilities, conduct argumentation based on evidence, and业目标：-2simulate based on selecting methods.Professional Ab在个人或团队环境下高效、负责地工作的能力。Abilitytowork effectivelyand responsiblyinan1-3ilityindividual or team context.展示个人和独立学习策略，扩展现有知识。培养信息检索和整合技能。Presentpersonal and1-4independent learning strategies to expand existing knowledge. Developing information retrieval andintegration skills(2）德培养责任感与团队协作精神，以及职业道德与行为规范。Cultivate a sense of responsibility and2-1育目标：teamwork spirit, as well as professional ethics and behavioral norms.Essential通过对模型的拟合优度分析以及残差分析，树立正确、严密的思维习惯。ThroughtheGoodnessof fit2-2Qualityanalysis and residual analysis of the model, establish correct and rigorous thinking habits三、教学内容Content(Topics)注：以中英文填写，各部分内容的表格可根据实际知识单元数量进行复制、扩展或缩减Note:Filled inboth CNand EN,extend or reducebased on theactual numbers ofknowledgeunit(1)理论教学Lecture知识单元序号：支撑教学目标：11-1, 1-2, 1-4,1-5Knowledge Unit NoSLOs Supported知识单元名称线性回归模型Unit TitleThe Linear Regression Model模型的建立知识点：Specifying the ModelKnowledge Delivery最小二乘估计Estimation withLeast Squares

二、教学目标 Subject Learning Objectives (SLOs) 注：毕业要求及指标点可参照悉尼学院本科生培养方案，可根据实际情况增减行数 Note: GA and index can be referred from undergraduate program in SSTC website. Please add/reduce lines based on subject. 整体目 标: Overall Objectiv e 高级统计建模是应用统计学专业的专业方向类课程，主要包括一元线性回归、多元线性回归、广义线性模 型、逻辑回归、顺序逻辑回归、多项逻辑回归、泊松回归等。本课程培养学生运用回归和分类模型解决复 杂问题的能力，对取得的成果进行评价和分析，并以各种方式向不同的受众（专家和非专家）简洁准确地 表达信息、推理和结论。 Advanced Statistical Modeling is a professional direction course for Applied Statistics majors, which mainly includes unitary linear regression, multiple linear regression, generalized linear model, logical regression, Ordinal Logistic Regression, Multinomial Logistic Regression, Poisson regression, etc. This course cultivates students' ability to use regression and classification models to solve complex problems, evaluate and analyze the results achieved, and express information, reasoning, and conclusions succinctly and accurately to different audiences (experts and non experts) in various ways. （1）专 业目标: Professi onal Ab ility 1-1 掌握各类回归模型的建模条件和方法，会对模型进行评价，并运用 SAS 在计算机实现。 Master the modeling conditions and methods of various regression models, be able to evaluate the models, and use SAS to implement them on a computer. 1-2 培养研究技能和解决问题的能力，在证据的基础上进行论证，并在选择方法的基础上进行模拟。 Cultivate research skills and problem-solving abilities, conduct argumentation based on evidence, and simulate based on selecting methods. 1-3 在个人或团队环境下高效、负责地工作的能力。Ability to work effectively and responsibly in an individual or team context. 1-4 展示个人和独立学习策略，扩展现有知识。培养信息检索和整合技能。Present personal and independent learning strategies to expand existing knowledge. Developing information retrieval and integration skills （2）德 育目标: Essential Quality 2-1 培养责任感与团队协作精神，以及职业道德与行为规范。Cultivate a sense of responsibility and teamwork spirit, as well as professional ethics and behavioral norms. 2-2 通过对模型的拟合优度分析以及残差分析，树立正确、严密的思维习惯。Through the Goodness of fit analysis and residual analysis of the model, establish correct and rigorous thinking habits. 三、教学内容 Content (Topics) 注：以中英文填写，各部分内容的表格可根据实际知识单元数量进行复制、扩展或缩减 Note: Filled in both CN and EN, extend or reduce based on the actual numbers of knowledge unit (1) 理论教学 Lecture 知识单元序号: Knowledge Unit No. 1 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 线性回归模型 The Linear Regression Model 知识点: Knowledge Delivery 模型的建立 Specifying the Model 最小二乘估计 Estimation with Least Squares

评估模型拟合Assessing Model Fit了解：模型的建立RecognizeSpecifying the Model学习目标：理解：最小二乘估计UnderstandEstimation with Least SquaresLearning Objectives掌握：评估模型拟合MasterAssessing Model Fit重点：最小二乘估计Key PointsEstimation with Least Squares难点：评估模型拟合Focal pointsAssessing Model Fit知识单元序号：支撑教学目标21-1,1-2,1-4,1-5SLOs SupportedKnowledge Unit No知识单元名称多元线性回归模型Unit TitleMultiple Linear Regression Model多元线性回归模型TheModel for MultipleLinearRegression知识点：最小二乘的矩阵形式Knowledge DeliveryOLSinmatrixform模型构建和解释Model building and interpretation了解:多元线性回归模型RecognizeThe Model for Multiple Linear Regression学习目标：理解：最小二乘的矩阵形式OLSinmatrixformLearning ObjectivesUnderstand掌握：模型构建和解释MasterModel building and interpretation重点：最小二乘的矩阵形式Key PointsOLS in matrix form难点：模型构建和解释Focal pointsModel building and interpretation知识单元序号支撑教学目标31-1, 1-2, 1-4, 1-5Knowledge Unit No.SLOs Supported知识单元名称广义线性建模框架Unit TitleGeneralized LinearModellingFramework指数分布族TheExponential FamilyofDistributions知识点：指数族的一般极大似然估计Knowledge DeliveryGeneral MLEfortheExponentialFamily广义线性建模框架

评估模型拟合 Assessing Model Fit 学习目标: Learning Objectives 了解: Recognize 模型的建立 Specifying the Model 理解: Understand 最小二乘估计 Estimation with Least Squares 掌握: Master 评估模型拟合 Assessing Model Fit 重点: Key Points 最小二乘估计 Estimation with Least Squares 难点: Focal points 评估模型拟合 Assessing Model Fit 知识单元序号: Knowledge Unit No. 2 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 多元线性回归模型 Multiple Linear Regression Model 知识点: Knowledge Delivery 多元线性回归模型 The Model for Multiple Linear Regression 最小二乘的矩阵形式 OLS in matrix form 模型构建和解释 Model building and interpretation 学习目标: Learning Objectives 了解: Recognize 多元线性回归模型 The Model for Multiple Linear Regression 理解: Understand 最小二乘的矩阵形式 OLS in matrix form 掌握: Master 模型构建和解释 Model building and interpretation 重点: Key Points 最小二乘的矩阵形式 OLS in matrix form 难点: Focal points 模型构建和解释 Model building and interpretation 知识单元序号: Knowledge Unit No. 3 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 广义线性建模框架 Generalized Linear Modelling Framework 知识点: Knowledge Delivery 指数分布族 The Exponential Family of Distributions 指数族的一般极大似然估计 General MLE for the Exponential Family 广义线性建模框架

TheGeneralizedLinearModellingFramework正态分布示例ExamplewiththeNormal Distribution了解：指数分布族The Exponential Family of DistributionsRecognize学习目标：理解：广义线性建模框架UnderstandThe Generalized Linear Modelling FrameworkLearning Objectives掌握：正态分布示例MasterExamplewiththeNormalDistribution重点：广义线性建模框架Key PointsThe Generalized Linear Modelling Framework难点：正态分布示例Focal pointsExample with the Normal Distribution知识单元序号：支撑教学目标：41-1, 1-2, 1-4, 1-5Knowledge Unit No.SLOs Supported逻辑回归导论知识单元名称Unit TitleIntroduction to Logistic Regression逻辑回归模型The Logistic Regression Model应用于二进制数据-分类x知识点：Applicationtobinarydata-categorical xKnowledge Delivery二进制数据的应用——连续xApplication to binary data - continuous x评估模型的拟合Assessing thefit of the model了解：逻辑回归模型RecognizeThe Logistic Regression Model学习目标：理解：应用于二进制数据-分类xUnderstandLearning ObjectivesApplication to binary data -categorical x掌握：评估模型的拟合MasterAssessing the fit of the model重点：应用于二进制数据-分类xKey PointsApplicationtobinarydata-categorical x难点：评估模型的拟合Focal pointsAssessing the fit of the model知识单元序号支撑教学目标51-1, 1-2, 1-4, 1-5Knowledge Unit No.SLOs Supported知识单元名称更复杂的Logistic回归模型Unit TitleMoreComplexLogisticRegression Models知识点：评估模型的适拟合Knowledge DeliveryAssessing the fit of the model

The Generalized Linear Modelling Framework 正态分布示例 Example with the Normal Distribution 学习目标: Learning Objectives 了解: Recognize 指数分布族 The Exponential Family of Distributions 理解: Understand 广义线性建模框架 The Generalized Linear Modelling Framework 掌握: Master 正态分布示例 Example with the Normal Distribution 重点: Key Points 广义线性建模框架 The Generalized Linear Modelling Framework 难点: Focal points 正态分布示例 Example with the Normal Distribution 知识单元序号: Knowledge Unit No. 4 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 逻辑回归导论 Introduction to Logistic Regression 知识点: Knowledge Delivery 逻辑回归模型 The Logistic Regression Model 应用于二进制数据-分类 x Application to binary data – categorical x 二进制数据的应用——连续 x Application to binary data – continuous x 评估模型的拟合 Assessing the fit of the model 学习目标: Learning Objectives 了解: Recognize 逻辑回归模型 The Logistic Regression Model 理解: Understand 应用于二进制数据-分类 x Application to binary data – categorical x 掌握: Master 评估模型的拟合 Assessing the fit of the model 重点: Key Points 应用于二进制数据-分类 x Application to binary data – categorical x 难点: Focal points 评估模型的拟合 Assessing the fit of the model 知识单元序号: Knowledge Unit No. 5 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 更复杂的 Logistic 回归模型 More Complex Logistic Regression Models 知识点: Knowledge Delivery 评估模型的适拟合 Assessing the fit of the model

具有多个类别的分类变量Categorical variableswithmultiplecategories多元逻辑回归Multiple Logistic Regression添加交叉项Adding Interaction Terms了解:具有多个类别的分类变量RecognizeCategorical variables with multiple categories理解：学习目标：多元逻辑回归Multiple Logistic RegressionLearning ObjectivesUnderstand掌握：添加交叉项MasterAdding Interaction Terms重点：多元逻辑回归Key PointsMultiple Logistic Regression难点：添加交叉项Focal pointsAdding Interaction Terms知识单元序号：支撑教学目标61-1, 1-2, 1-4, 1-5SLOs SupportedKnowledge Unit No.有序回归知识单元名称Unit TitleOrdinal Regression有序逻辑回归模型TheOrdinal LogisticRegressionModel有序数据的应用一一分类和连续x知识点：Applicationtoordinaldata-categorical&continuousxKnowledge Delivery评估模型的拟合性Assessing thefit ofthemodel扩展线性预测Extending the linear predictor了解:有序逻辑回归模型RecognizeTheOrdinal LogisticRegressionModel理解：有序数据的应用一一分类和连续x学习目标UnderstandApplication to ordinal data-categorical & continuous xLearning Objectives评估模型的拟合性掌握：Assessing the fit of the modelMaster扩展线性预测Extendingthelinearpredictor重点：月——分类和连续x有序数据的应用Key PointsApplicationtoordinaldata-categorical&continuousx难点：评估模型的拟合性Focal pointsAssessingthefitofthemodel

具有多个类别的分类变量 Categorical variables with multiple categories 多元逻辑回归 Multiple Logistic Regression 添加交叉项 Adding Interaction Terms 学习目标: Learning Objectives 了解: Recognize 具有多个类别的分类变量 Categorical variables with multiple categories 理解: Understand 多元逻辑回归 Multiple Logistic Regression 掌握: Master 添加交叉项 Adding Interaction Terms 重点: Key Points 多元逻辑回归 Multiple Logistic Regression 难点: Focal points 添加交叉项 Adding Interaction Terms 知识单元序号: Knowledge Unit No. 6 支撑教学目标: SLOs Supported 1-1，1-2，1-4，1-5 知识单元名称 Unit Title 有序回归 Ordinal Regression 知识点: Knowledge Delivery 有序逻辑回归模型 The Ordinal Logistic Regression Model 有序数据的应用——分类和连续 x Application to ordinal data – categorical & continuous x 评估模型的拟合性 Assessing the fit of the model 扩展线性预测 Extending the linear predictor 学习目标: Learning Objectives 了解: Recognize 有序逻辑回归模型 The Ordinal Logistic Regression Model 理解: Understand 有序数据的应用——分类和连续 x Application to ordinal data – categorical & continuous x 掌握: Master 评估模型的拟合性 Assessing the fit of the model 扩展线性预测 Extending the linear predictor 重点: Key Points 有序数据的应用——分类和连续 x Application to ordinal data – categorical & continuous x 难点: Focal points 评估模型的拟合性 Assessing the fit of the model