《金融数据分析》教学大纲The Course Outline of Financial DataAnalysis课程编号:151222B课程类型:专业选修课总学时:32 讲课学时:16 实验(上机)学时:16学分:2适用对象:金融学(金融经济实验班)先修课程:计量经济学、微观经济学、宏观经济学、概率论与数理统计、线性代数、微积分Course Code:Course Type: Discipline Elective CoursePeriods: 32 Lecture: 16 Experiment (Computer): 16Credits: 2Applicable Subjects: Finance(Finance and Economics Experiment Class)Prerequisite Courses: Econometrics, Microeconomics, Macroeconomics, Probability and Statistics, Linear Algebra, Calculus一、课程简介本课程是面向金融学、经济学和管理学相关专业的高年级本科生开设的学科专业选修课程,主要介绍应用于金融数据分析中的经典计量方法,并注重培养学生的实际操作能力。
Financial data analysis is an elective course for advanced undergraduate students majored in finance, economics and management. In this course, we not only introduce basic econometric methods applied to financial data, but also train students’ practical skills of handling financial data.二、教学目标本课程包括以下教学目标:(1)使学生能运用所学的金融计量理论分析和解释实际金融数据;(2)着重为对金融理论进行实证分析提供所需的估计和检验方法;(3)使学生运用计算机软件(以R语言为主)进行金融数据研究。
This course has the following teaching goals: (1) equipping students with econometric tools to analyze and explain financial data, (2) focusing on the estimation and inferential methods used in empirical analysis of finance theory, and (3) enabling students to implement these techniques using computer software, primarily R programming language.二、教学基本要求这门课程主要讲授如何利用海量金融数据对金融理论进行实证分析,以及在处理实际金融数据时所需的计量方法和计算机技术。
因此,在教学内容的讲授过程中,授课老师需要做到理论与实践并重。
理论学习包括金融数据概述,平稳时间序列分析,非平稳性时间序列分析,波动率分析,股票回报率预测,等。
在时间允许的情况下,我们还将介绍,多维向量自回归模型, 以及协整和误差纠正模型。
对于上述所有的模型的分析,本课程提供相应的实际金融数据,以及完整的R程序与讲解说明文件。
本课程教学方法将以课堂讲授为主。
由于课程内容难度相对较大,我们鼓励学生课前预习,课上积极参与讨论,并课后复习和独立完成作业。
R语言的学习以课堂展示和学生上机实际操作的方式完成。
课程考核由考勤、作业、研究项目、和期末考试组成。
考核成绩为百分制,各项分数分配见表(一)。
表(一):分数分配方式出勤10%作业30%期末闭卷考试30%研究项目30%The main content of this course is to explain how to handle a huge amount of financial data and to conduct empirical analysis based on financial theories. Thus, this course will weight equally on both learning econometric theories and real applications. Theoretical study will mainly cover such topics as introduction to financial data, stationary time series analysis, non-stationary time series analysis, volatility analysis, and stock return prediction. If time permits, we will also learn vector autoregressive models (V AR) and cointegration and error correction models. All the models are accompanied with real-data examples with complete R programs and tutorials.The basic teaching strategy of this course mainly involves in-class lectures. Due to the difficulty of the course, we encourage students to browse the assigned reading materials before class, to proactively participate discussions in class, and to peruse lecture notes and independently complete homework after class. Students will learn R through in-class instructions and practice.The methods of the course evaluation include attendance, homework, project work, and final exam. The grade distribution of each component within one hundred percentage points is presented in Table 1Table 1: The grade distributionAttendance 10%Assignments 30%Final Exam (closed) 30%Project Work 30%三、各教学环节学时分配教学课时分配(Class Schedule)四、教学内容第一章金融市场数据的特征第一节数据:金融数据的实证特征对金融建模很重要第二节概括性的统计量第三节分位数相关的分析1.在险价值2.期望损失3.密度预测第四节实证分析1.有效市场假说2.方差比检验教学重点、难点:在险价值和期望损失的计算课程的考核要求:计算在险价值和期望损失,用计量软件检验有效市场假说第二章平稳时间序列第一节ARMA过程1.数据2.模型性质第二节估计,检验和模型选择1.MLE估计方法2.统计量3.ARMA模型选择第三节预测1.预测过程2.预测评估第四节实证分析教学重点、难点:ARMA过程模型性质和预测课程的考核要求:分析ARMA过程模型性质和预测,用计量软件预测ARMA过程第三章非平稳时间序列第一节单位根过程和确定性时间序列1.数据2.单位根过程的模型性质3.含有确定性时间序列模型的性质第二节估计,检验和模型选择1.MLE估计方法2.单位根检验3.滞后阶数的选择第三节预测1.预测过程2.预测评估第四节实证分析1.泡沫检验2.泡沫起始时间的估计教学重点、难点:单位根检验和泡沫检验课程的考核要求:分析单位根过程的模型性质,用计量软件进行单位根检验第四章波动率第一节资产价格过程的波动聚类第二节自回归条件异方差模型(ARCH)1.模型性质2.估计方法3.模型选择第三节广义自回归条件异方差模型(GARCH)1.模型性质2.估计方法3.模型选择第四节实证分析教学重点、难点:ARCH和GARCH模型的估计课程的考核要求:用ARCH和GARCH模型来分析现实金融数据的波动性第五章股票回报率预测第一节股票回报率预测(i)3.预测回归4.长期预测回归第二节股票回报率预测(ii)5.资本资产定价模型6.多因子模型第三节实证分析教学重点、难点:预测回归和多因子模型课程的考核要求:用计量软件检验股票收益的可预测性第六章多元时间序列建模第一节向量自回归过程1.数据2.向量自回归过程的模型性质3.估计,检验和模型选择4.脉冲响应函数分析5.预测方差分解第二节伪回归和协整模型1.伪回归模型2.协整模型第三节向量误差修正模型1.模型性质2.模型估计教学重点、难点:向量自回归过程的模型性质和协整关系检验课程的考核要求:分析向量自回归过程的模型性质,用计量软件进行协整关系检验Chapter 1 The Characteristics of Financial Market DataSection 1 The Data: The empirical characteristics of financial data are important for building financial models.Section 2 Summary StatisticsSection 3 Percentiles related statistics1.Value-at-Risk2.Expected Shortfall3.Density Forecasting EstimationSection 4 Empirical Analysis1.Effcient Market Hypothesis2.Variance Ratio TestsKey and Difficult Points: calculation of Value-at-Risk and Expected Shortfall Evaluation Requirements: calculation of Value-at-Risk and Expected Shortfall, and testing the Effcient Market Hypothesis by econometric softwareChapter 2 Stationary Time SeriesSection 1 ARMA processes1.Data2.PropertiesSection 2 Estimation, Inference and Model Selection1.Estimation by MLE2.Test Statistics3.ARMA Model SelectionSection 3 Forecasting1.Forecasting Procedure2.Forecasting EvaluationSection 4 Empirical AnalysisKey and Difficult Points: the properties of ARMA process and forecasting using ARMA modelEvaluation Requirements: the properties of ARMA process, and forecasting using ARMA model by econometric softwareChapter 3 Nonstationary Time SeriesSection 1 Unit Root and Deterministic Trends1.Data2.Properties of Unit Root Model3.Properties of Model with Deterministic TrendsSection 2 Estimation, Inference and Model Selection1.Estimation by MLE2.Unit Root Testg-length Model SelectionSection 3 Forecasting1.Forecasting Procedure2.Forecasting EvaluationSection 4 Empirical Analysis1.Bubble Test2.Data Stamping of BubblesKey and Difficult Points: Unit root test and Data Stamping of BubblesEvaluation Requirements: Properties of Unit Root Model, and test unit root by econometric softwareChapter 4 VolatilitySection 1 V olatility Clustering of Asset Price ProcessesSection 2 Autoregressive Conditional Heteroscedasticity Model(ARCH)1.Properties2.Estimation3.Model SelectionSection 3 Generalized Autoregressive Conditional Heteroscedasticity Model(GARCH)1.Properties2.Estimation3.Model SelectionSection 4 Empirical AnalysisKey and Difficult Points: estimation of ARCH and GARCH modelsEvaluation Requirements:apply ARCH and GARCH models to real data in financeChapter 5 Stock Return PredictabilitySection 1 Stock Return Predictability (i)1.Predictive Regression2.Long-horizon Predictive RegressionSection 2 Stock Return Predictability (ii)1.Capital Asset Pricing Model (CAPM)2.Multifactor ModelsSection 3 Empirical AnalysisKey and Difficult Points: Predictive Regression and Multifactor ModelsEvaluation Requirements: Test stock return predictability by PredictiveRegression using econometric softwareChapter 6 Multivariate Time SeriesSection 1 Vector Autoregressive Models1.Data2.Properties of Vector Autoregressive Models3.Estimation, Inference and Model Selection4.Impulse Response Analysis5.Forecast error variance decompositionsSection 2 Spurious Regression and Cointegration1.Spurious Regression2.CointegrationSection 3 Vector Error Correction Model1.Properties2.EstimationKey and Difficult Points: Vector Autoregressive Models and CointegrationEvaluation Requirements: Properties of Vector Autoregressive Models and Cointegration test by econometric software五、其它授课老师应根据学生的能力适当调节课程进度和内容Instructor should adjust the course schedule and content according to the capability of students.六、主要参考书[1] Walter Enders. Applied Econometric Time Series, 4th edition. Wiley. 2014[2] Ruey S. Tsay.An Introduction to Analysis of Financial Data with R. Wiley.2012执笔人:田峥教研室主任:系教学主任审核签名:11。