《計量經(jīng)濟學(xué)》ch-01-wooldridge-5e-教學(xué)課件

1、Chapter 1 The Nature of Econometrics and Economic DataWooldridge: Introductory Econometrics: A Modern Approach, 5eInstructed by professor Yuan, Huiping第1頁，共26頁。Chapter 1 The Nature of Econometrics and Economic Data1.1 What Is Econometrics?1.2 Steps in Empirical Economic Analysis1.3 The Structure of

2、Economic Data1.4 Causality and Ceteris Paribus in Econometric AnalysisThe EndAppendix A: Basic Mathematical Tools第2頁，共26頁。Chapter 1 The Nature of Econometrics and Economic DataWhat is econometrics?Econometrics = use of statistical methods to analyze economic dataEconometricians typically analyze non

3、experimental dataTypical goals of econometric analysisEstimating relationships between economic variablesTesting economic theories and hypothesesForecasting economic variablesEvaluating and implementing government and business policyChapterEnd1.1 What Is Econometrics?第3頁，共26頁。Steps in econometric an

4、alysis1) Economic model (this step is often skipped)2) Econometric modelEconomic modelsMaybe micro- or macromodelsOften use optimizing behaviour, equilibrium modeling, Establish relationships between economic variablesExamples: demand equations, pricing equations, Chapter 1 The Nature of Econometric

5、s and Economic DataChapterEnd1.2 Steps in Empirical Economic Analysis (1/5)第4頁，共26頁。Economic model of crime (Becker (1968)Derives equation for criminal activity based on utility maximizationFunctional form of relationship not specifiedEquation could have been postulated without economic modelingHour

6、s spent incriminal activitiesWage“ of cri-minal activitiesWage for legalemploymentOtherincomeProbability ofgetting caughtProbability ofconviction ifcaughtExpectedsentenceAgeChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.2 Steps in Empirical Economic Analysis (2/5)第5頁，共26頁。Model of

7、 job training and worker productivityWhat is effect of additional training on worker productivity?Formal economic theory not really needed to derive equation:Other factors may be relevant, but these are the most important (?)Hourly wageYears of formaleducationYears of work-force experienceWeeks spen

8、tin job trainingChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.2 Steps in Empirical Economic Analysis (3/5)第6頁，共26頁。Econometric model of criminal activityThe functional form has to be specifiedVariables may have to be approximated by other quantitiesMeasure of cri-minal activityWa

9、ge for legalemploymentOtherincomeFrequency of prior arrestsFrequency of convictionAverage sentencelength after convictionAgeUnobserved deter-minants of criminal activitye.g. moral character, wage in criminal activity,family background Chapter 1 The Nature of Econometrics and Economic DataChapterEnd1

10、.2 Steps in Empirical Economic Analysis (4/5)第7頁，共26頁。Econometric model of job training and worker productivityMost of econometrics deals with the specification of the error Econometric models may be used for hypothesis testingFor example, the parameter represents effect of training on wageHow large

11、 is this effect? Is it different from zero?Hourly wageYears of formaleducationYears of work-force experienceWeeks spentin job trainingUnobserved deter-minants of the wagee.g. innate ability,quality of education,family background Chapter 1 The Nature of Econometrics and Economic DataChapterEnd1.2 Ste

12、ps in Empirical Economic Analysis (5/5)第8頁，共26頁。Different kinds of economic data setsCross-sectional dataTime series dataPooled cross sectionsPanel/Longitudinal dataEconometric methods depend on the nature of the data usedUse of inappropriate methods may lead to misleading resultsChapter 1 The Natur

13、e of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (1/10)第9頁，共26頁。Cross-sectional data setsSample of individuals, households, firms, cities, states, countries, or other units of interest at a given point of time/in a given periodCross-sectional observations are more or l

14、ess independentFor example, pure random sampling from a populationSometimes pure random sampling is violated, e.g. units refuse to respond in surveys, or if sampling is characterized by clusteringCross-sectional data typically encountered in applied microeconomicsChapter 1 The Nature of Econometrics

15、 and Economic DataChapterEnd1.3 The Structure of Economic Data (2/10)第10頁，共26頁。Observation numberHourly wageIndicator variables(1=yes, 0=no)Cross-sectional data set on wages and other characteristicsChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (3

16、/10)第11頁，共26頁。Cross-sectional data on growth rates and country characteristicsAdult secondaryeducation ratesGovernment consumtionas percentage of GDPGrowth rate of realper capita GDPChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (4/10)第12頁，共26頁。Tim

17、e series dataObservations of a variable or several variables over timeFor example, stock prices, money supply, consumer price index, gross domestic product, annual homicide rates, automobile sales, Time series observations are typically serially correlatedOrdering of observations conveys important i

18、nformationData frequency: daily, weekly, monthly, quarterly, annually, Typical features of time series: trends and seasonalityTypical applications: applied macroeconomics and financeChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (5/10)第13頁，共26頁。Tim

19、e series data on minimum wages and related variablesUnemployment rateAverage coverage rateAverage minimum wage for given yearGross nationalproductChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (6/10)第14頁，共26頁。Pooled cross sectionsTwo or more cross

20、sections are combined in one data setCross sections are drawn independently of each otherPooled cross sections often used to evaluate policy changesExample:Evaluate effect of change in property taxes on house pricesRandom sample of house prices for the year 1993A new random sample of house prices fo

21、r the year 2019Compare before/after (1993: before reform, 2019: after reform)Chapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (7/10)第15頁，共26頁。Pooled cross sections on housing pricesNumber of bathroomsSize of housein square feetProperty taxBefore refo

22、rmAfter reformChapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Economic Data (8/10)第16頁，共26頁。Panel or longitudinal dataThe same cross-sectional units are followed over timePanel data have a cross-sectional and a time series dimensionPanel data can be used to accou

23、nt for time-invariant unobservablesPanel data can be used to model lagged responsesExample:City crime statistics; each city is observed in two yearsTime-invariant unobserved city characteristics may be modeledEffect of police on crime rates may exhibit time lagChapter 1 The Nature of Econometrics an

24、d Economic DataChapterEnd1.3 The Structure of Economic Data (9/10)第17頁，共26頁。Two-year panel data on city crime statisticsEach city has two time series observationsNumber of police in 1986Number of police in 1990Chapter 1 The Nature of Econometrics and Economic DataChapterEnd1.3 The Structure of Econo

25、mic Data (10/10)第18頁，共26頁。Causality and the notion of ceteris paribusMost economic questions are ceteris paribus questionsIt is important to define which causal effect one is interested inIt is useful to describe how an experiment would have to be designed to infer the causal effect in questionDefin

26、ition of causal effect of on : How does variable change if variable is changedbut all other relevant factors are held constant“ChapterEnd1.4 Causality and Ceteris Paribus in Econometric Analysis (1/6)Chapter 1 The Nature of Econometrics and Economic Data第19頁，共26頁。Causal effect of fertilizer on crop

27、yieldBy how much will the production of soybeans increase if one increases the amount of fertilizer applied to the ground“Implicit assumption: all other factors that influence crop yield such as quality of land, rainfall, presence of parasites etc. are held fixedExperiment:Choose several one-acre pl

28、ots of land; randomly assign different amounts of fertilizer to the different plots; compare yieldsExperiment works because amount of fertilizer applied is unrelated to other factors influencing crop yieldsChapterEnd1.4 Causality and Ceteris Paribus in Econometric Analysis (2/6)Chapter 1 The Nature

29、of Econometrics and Economic Data第20頁，共26頁。Measuring the return to educationIf a person is chosen from the population and given another year of education, by how much will his or her wage increase? “Implicit assumption: all other factors that influence wages such as experience, family background, in

30、telligence etc. are held fixedExperiment:Choose a group of people; randomly assign different amounts of eduction to them (infeasable!); compare wage outcomesProblem without random assignment: amount of education is related to other factors that influence wages (e.g. intelligence)ChapterEnd1.4 Causal

31、ity and Ceteris Paribus in Econometric Analysis (3/6)Chapter 1 The Nature of Econometrics and Economic Data第21頁，共26頁。Effect of law enforcement on city crime levelIf a city is randomly chosen and given ten additional police officers, by how much would its crime rate fall? “Alternatively: If two citie

32、s are the same in all respects, except that city A has ten more police officers, by how much would the two cities crime rates differ?“Experiment:Randomly assign number of police officers to a large number of citiesIn reality, number of police officers will be determined by crime rate (simultaneous d

33、etermination of crime and number of police) ChapterEnd1.4 Causality and Ceteris Paribus in Econometric Analysis (4/6)Chapter 1 The Nature of Econometrics and Economic Data第22頁，共26頁。Effect of the minimum wage on unemploymentBy how much (if at all) will unemployment increase if the minimum wage is increased by a certain amount (holding other things fixed)? “Experiment:Government randomly chooses minimum wage each year and observes unemployment outcomesExperiment will work because level of minimum wage is unrelated to other factors determining unemploymentIn reality, the level of