Outline of lectures
PART I: Regression analysis
Lecture 1: The aim of this lecture is to provide an introduction to regression analysis using the CLRM.
By the end of it students should be able to:
• Explain the traditional methodology in econometrics
• Differentiate the notions of regression, causation and correlation
• Recognize the terminology and notation of regression analysis
• Compare the population regression function with the sample regression function
• Explain why the error term is important in regression analysis
Main text reading: DG Chapters 1 and 2.

Lecture 2: The aim of this lecture is to introduce various concepts and assumptions related to the estimation of a bivariate regression model.
By the end of it students should be able to:
• Use the method of ordinary least squares (OLS) to compute estimators and their standard errors in the bivariate regression model
• Explain the assumptions underlying OLS in the bivariate regression model
• Explain the Gauss-Markov theorem
• Compute goodness of fit measures for the bivariate regression model
Main text reading: DG Chapter 3.

Lecture 3: In this lecture, we will analyse issues related to the estimation of a multivariate regression model.
By the end of it students should be able to:
• Use the OLS method to compute estimators and their standard errors in the multivariate regression model
• Explain the assumptions underlying OLS in the multivariate regression model
• Illustrate the properties of OLS estimators
• Compute goodness of fit measures for the multivariate regression model
Main text reading: DG Chapter 7.

Lecture 4: In this lecture we will explain the matrix approach to multivariate regression modelling.
By the end of it students should be able to:
• Write the matrix representation of the k-variable regression model
• Explain the OLS assumptions in matrix format
• Compute the vector of OLS estimators and their variance-covariance matrix
• Illustrate the properties of the vector of OLS estimators
Main text reading: DG Appendix C.

PART II: Interval estimation and hypothesis testing
Lecture 5: The aim of this lecture is to introduce the concepts of interval estimation and hypothesis testing using the bivariate regression model.
By the end of it students should be able to:
• Explain the interval estimator
• Construct a confidence interval
• Perform hypothesis testing using the confidence interval, and t-test of significance approaches
Main text reading: DG Chapter 5.

Lecture 6: The aim of this lecture is to explain how various hypotheses can be tested within the multivariate regression model.
By the end of it students should be able to:
• Test hypotheses about individual coefficients
• Test the equality of two coefficients
• Test joint hypotheses
• Test linear restrictions
Main text reading: DG Chapter 8.

PART III: Relaxing the assumptions of the classical linear regression model
Lecture 7: The aim of this lecture is to analyse the problem of heteroskedasticity.
By the end of it students should be able to:
• Summarize the sources of heteroskedasticity
• Illustrate the consequences of heteroskedasticity for OLS estimators and their variances
• Detect heteroscedasticity using formal and informal methods
• Apply remedial measures for heteroskedasticity
Main text reading: DG Chapter 11.
LECTURE_7_BE.pdf
Lecture 8: The aim of this lecture is to analyse the problem of autocorrelation.
By the end of it students should be able to:
• Summarize the sources of autocorrelation
• Illustrate the consequences of autocorrelation for OLS estimators and their variances
• Detect autocorrelation using formal and informal methods
• Apply remedial measures for autocorrelation
Main text reading: DG Chapter 12.
LECTURE_8_BE_rev.pdf
Lecture 9: In this lecture we will analyse various issues related to multicollinearity, non-linearity and non-stationarity.
By the end of it students should be able to:
• Summarize the sources of multicollinearity
• Illustrate the consequences of multicollinearity for OLS estimators and their variances
• Distinguish between intrinsically linear and non-linear regression models

• Contrast stationary with non-stationary processes
Main text reading: DG Chapters 10, 14, 21.
LECTURE_9_BE.pdf
• Lecture 10: In this lecture we will analyse some academic papers that have applied regression analysis to examine economic and/or finance-related hypotheses of interest.
Main text reading: TBA.

Labs outline
Lab 1: The aim of this lab is to provide an introduction to applied regression analysis using appropriate econometric software.
By the end of it students should be able to:
• Input and save data files
• Perform data transformations and preliminary data analysis
• Perform graphical analysis
• Specify a linear regression model and compute OLS estimates

Lab 2: The aim of this lab is to show how to conduct hypothesis testing using appropriate econometric software.
By the end of it students should be able to:
• Perform hypothesis testing in a bivariate regression model
• Perform hypothesis testing in a multivariate regression model

Lab 3: The aim of this lab is to demonstrate additional regression models and explain the regression’s diagnostics.
By the end of it students should be able to:
• Specify and estimate an autoregressive model
• Interpret the regression’s diagnostics
• Test for structural stability using sub-sample regressions
• Specify and estimate a dummy variable model

Lab 4: TBA

lecture1-9的讲义合集

LAB讲义合集