中级计量经济学：Chapter 1 Introduction of Econometrics.ppt

,Click to edit Master title style,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,Introductory Econometrics for Finance Chris Brooks 2008,Econometrics,1.1 What is Econometrics?,1.2 The Nobel Prize&Econometrics,1.3 The Methodology of Econometrics,1.4 How to study Econometrics?,1.5 An econometric case study,Chapter 1 Introduction of Econometrics,SUIBE CHEN HUA,“Econometrics is the study of the application of statistical methods to the analysis of economics phenomena.”,“the application of mathematical and statistical techniques to economic data and problems.”,“The application of statistical and mathematical methods in the field of economics to describe the numerical relationships between key economic forces such as capital,interest rates,and labor.”,1.1 What is Econometrics?,“Application of statistical techniques in evaluation and testing of economic theories.”,“Application of mathematical and statistical techniques to economics in the study of problems,the analysis of data,and the development and testing of theories and models.”,“The application of statistical and mathematical theories to economics for the purpose of testing hypotheses and forecasting future trends.”,“Econometrics is what econometricians do.”,Econometricians are now routinely employed in government and business to estimate and forecast phenomena,such as,(1)price and cost elasticity,(2)production and cost functions,(3)demand functions for goods and services,etc.,We can see that econometric forecasting is a growth industry.,What econometricians do?,Frist,and foremost,they are economists,capable of utilizing economic theory to improve their empirical analyses of problem they address.,At times they are mathematicians,formulating economic theory in ways that make it appropriate for statistic testing.,At times they are statisticians,applying their skills to the development of statistical techniques appropriate to the empirical problems,and spending hours with the computers trying to estimate economic relationships to predict economic events.,What econometricians do?,Econometrics is a bridge that connects economics,statistics and mathematics.,It is a very useful and popular tool in the study of practical problems in many areas.,Economics,Statistics,Mathematics,Three key words,Economics,Mathematics,Statistics,Econometrics,Econometrics is a perfect combination within economics,mathematics and statistics.,Its a bridge to connect with economic theories and statistics.,1.2 The Nobel Prize&Econometrics,By 2014,a total of 81 economists won the Nobel Prize in economics.,14,of them directly because of the contribution to the development of econometrics,：,1969 R.Frish&J.Tinbergen,1973 W.Leotief,1980 L.R.Klein,1984 R.Stone,1989 T.Haavelmo,2000 J.J.Heckman&D.L.McFadden,2003 Cliver W.J.Granger&Robert F.Engle,2005 Robert J.Aumann&Thomas C.Schelling,2011 Thomas J.Sargent&Christopher A.Sims,17,of them served as Econometric Society Council,41,or so in the award-winning results in the application of econometrics.,Nobel Prizes in Economics,（,1969,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 1969,for having developed and applied dynamic models for the analysis of economic processes,Ragnar Frisch,Norway,Oslo University,Oslo,Norway,1895-1973,Jan Tinbergen,the Netherlands,The Netherlands School of Economics,Rotterdam,The Netherlands,1903-1994,Nobel Prizes in Economics,（,1973,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 1973,for the development of the input-output method and for its application to important economic problems,Wassily Leontief,USA,Harvard University,Cambridge,MA,USA,1906-1999,Nobel Prizes in Economics,（,1980,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 1980,for the creation of econometric models and the application to the analysis of economic fluctuations and economic policies,Lawrence R.Klein,USA,University of Pennsylvania,Philadelphia,PA,USA,1920-,Nobel Prizes in Economics,（,1984,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 1984,for having made fundamental contributions to the development of systems of national accounts and hence greatly improved the basis for empirical economic analysis,Richard Stone,Great Britain,Cambridge University,Cambridge,Great Britain,1913-1991,Nobel Prizes in Economics,（,1989,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 1989,for his clarification of the probability theory foundations of econometrics and his analyses of simultaneous economic structures,Trygve Haavelmo,Norway,University of Oslo,Oslo,Norway,1911-1999,Nobel Prizes in Economics,（,2000,）,James J Heckman Daniel L McFadden,USA USA,University of Chicago University of California,Chicago,IL,USA Berkeley,CA,USA,1944-1937-,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 2000,for his development of theory and methods for analyzing selective samples,for his development of theory and methods for analyzing discrete choice,Nobel Prizes in Economics,（,2003,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 2003,for methods of analyzing economic time series with time-varying volatility(ARCH),for methods of analyzing economic time series with common trends(cointegration),Robert F.Engle,USA,New York University,New York,NY,USA,1942,Clive W.J.Granger,United Kingdom,University of California San Diego,CA,USA,1934,Nobel Prizes in Economics,（,2005,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 2005,“Game Theory and Decision-making theory,“Game Theory,Robert J.Aumann,Israel&USA,Hebrew University of Jerusalem,1930-,Thomas C.Schelling,USA,Harvard University,1921-,Nobel Prizes in Economics,（,2011,）,The Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel 2011,“the expected role in macroeconomic models,dynamic economic theory and time series analysis,“Vector autoregression method to analyze how the economy by temporary changes in economic policy and other factors,（,VAR),THOMAS J.SARGENT,USA,New York University,New York,NY,USA,1943-,Christopher A.Sims,USA,Princeton University,1942-,The core ideas of econometrics:,make economic models,The core method of econometrics:,regression analysis,1.3 The Methodology of Econometrics,The steps of make model,(1)questions,(2)establish the independent variables and the dependent variables,(3)Data collection,(4)modeling,(5)parameter estimation,(6)the model test,(7)modify and analysis the problem by using model,(8)predict,Regression,Economic Implication of Regression,:,through the surface of things,phenomena and problems,get to the root and find the most important cause,affecting factors that impact.,Regression method,:,theoretical approach,least squares estimation(OLS),The actual operation,:through software(EVIEWS,STATA,SPSS,EXCEL,etc.),Econometrics will be a strong instrument.,We looked econometrics as an applicable tool,you should be able to use it to resolve the practical problems in your life and work.,Computer is a very good assistant,we can use software to estimate the models and testing.,1.4 How to study Econometrics,Descriptive statistics,Probability,Random variables,Probability distributions,Expectation and variance,Sampling,Estimation,Statistic inference and testing,What you need from Statistics?,Why you need Econometrics,Theoretical learning needs,The need for further study,Practical application,There is a case from Labor Economics,Question:,the economic situation will affect people entering the labor market decision yet?That is,if the economic situation will affect on peoples willingness to work?,1.5 A case used econometrics,Assuming a rate of unemployment(Unemployment Rate,UNR)to measure the economic situation,with the labor force participation rate(Labor Force Participation Rate,LFPR)to measure the labor force participation.,the data released by the government on time,then how to answer the above questions?,We would like to use the idea of econometric analysis.,discouraged-worker hypothesis,added-worker hypothesis,the labor force participation rates increase or decrease depending on the contrast between discouraged-worker and added-worker.,(1)Theory or Hypothesis Statement,Dependent variable:,the labor force participation rate(LFPR),Independent variables,Unemployment Rate(UNR),(2)independent variables and dependent variable,Types of statistical data,：,Time-serial data,cross-sectional data,pooled data(panel data),This case we use time-serial data,(3)Data collection,The,first,one is,time series data,.This is a data set of one variable over a time period.It consists of repeated observations through time of entities,with fixed intervals between the observations.Usually there are yearly data,quarterly data,monthly data and daily data.,The yearly data,quarterly data and monthly data would be used in macroeconomics,e.g.,the yearly data on Chinese GDP from 1990 to 2012.,The daily data is regularly used in financial research,e.g.,the daily average price of a stock for the past 30 weekdays.,The,second,one is,cross-sectional data,.This is the different observations of the same variable at one moment in time.The units of observation may be individuals,households,enterprises,countries,or any set of elements that are sufficiently similar in nature.,For example,the GDP data of 31 different Chinese provinces in 2012.,For another example,the price of 50 different stocks on the same day.,The,third,one is,panel data,.That includes both time series data and cross-sectional data,consisting of repeated observations on the same elements through time.,An example is the GDP of the 31 Chinese provinces from 1990 to 2012.This data set consists of yearly time series data and cross-sectional data in every year.,Another example is the daily price of 50 different stocks for the past 30 weekdays.,Table 1-1,U.S.Civilian Labor Force Participation Rate(CLFPR),CivilianUnemployment Rate(CUNR),and Real Average Hourly Earnings(AHE82)*for the years 1980-2002.,Firstly,we use scatter diagram to find the relationship between the CLFPR and CUNR.,(4)Make model,Figure 1-1,Regression plot for civilian labor force participationrate(%)and civilian unemployment rate(%).,Secondly,we create a simple mathematical model:,CLFPR=B,1,+B,2,CUNR （1-1）,That is a simple linear function.,B1 and B2 are the parameters of this linear function.,B1 is the intercept,and it means the value of CLFPR when CUNR is equal to zero.,B2 is the slope,it means the change of each unit CUNR caused to the change of CLFPR.,Thirdly,we create an econometrics model:,CLFPR=B,1,+B,2,CUNR+u （1-2）,u is,random error term,(,error term),We used least squares method(OLS)to obtain the following:,CLFPR=69.9355-0.6458CUN(1-3),(5)parameter estimation,coe,fficient Std.Error t-Statistic Prob.,B169.93547 0.987646 70.810250.0000,B2-0.6458230.141562 -4.5621160.0004,R-squared 0.581157 Mean dependent var,65.50588,Adjusted R-squard 0.553234 S.D.dependent var,1.115499,S.E.of regression 0.745606 Akaike info criterion,2.360892,Sum squared resid 8.338928 Schwarz criterion,2.458918,Log likelihood -18.06759 Durbin-Watson stat,0.572261,CLFPR=69.9355-0.6458CUNR,se=(,0.987646,)(,0.141562,),t=(,70.81025,)(,-4.562116,),p value,=,(,0.0000,)(,0.0004,),R,2,=0.58,d.f.=15,Economic testing,Statistical testing,Econometric testing,(6)Model testing,In front of our model,we did not considered the other factors those would affected the labor market,so let us do further analysis.,CLFPR=B,1,+B,2,CUNR+B,3,AHE82+u （1-4）,Through OLS:,CLFPR=97.9-0.446CUNR-3.86AHE82,（1-5）,(7)further discussion and modify model,From the model we can got a conclusion that CUNR was indeed effected the CLFPR,there is negative correlation between these.,When CUNR improved 1%,the CLFPR will decreased about 0.65%.,So,the conclusion supported the discouraged-worker hypothesis.,And we can use the model forecast.,(8)Analyze problems and predict,