Intermediate Econometrics

Description:

Intermediate Econometrics

The typical introductory econometrics course is mostly confined to the Classical Linear Regression Model. It may also cover a few advanced topics, but with eclectic choice and superficial treatment. This course aims to build a solid, comprehensive understanding of the use of the regression model when one progresses from the CLRM, with its strong and unrealistic assumptions, and addresses the issues that researchers encounter in practice.

Who is this course aimed at?
This course is designed for those who have already taken an introductory course covering the CLRM and need to develop their skills to the next level. The course is not suitable for those whose only exposure to econometrics has been as part of a financial statistics course.

Course benefitsStudents will have an advanced understanding of the use of the regression model beyond the CLRM. They should then be able to apply econometric techniques to real-world issues. The course also serves as preparation for an MSc level econometrics course.

Prerequisites
At least one semester of mathematical statistics with a serious analytical treatment of estimation and inference, and at least one semester of multivariate calculus, both passed at a respectable standard. The CLRM foundation is a mandatory prerequisite and applicants should be careful to provide evidence of meeting it.

Course outline

The course assumes that participants will already have a sound understanding of the CLRM. After a brief review of the latter, it will cover the following topics:

• stochastic regressors
• measurement error
• instrumental variables
• simultaneous equations estimation

Next, it will cover binary choice models:

• linear probability model and its shortcomings
• logit model
• probit model
• tobit model
• sample selection bias

The course will then treat issues related to regressions with time series data:

• regressions with lagged variables
• autoregressive distributed lag - ADL(p,q) - models
• detailed treatment of issues relating to the ADL(1,0) model
• consequences of the violation of the assumption of noncontemporaneous independence of regressors and disturbance term
• autocorrelation: consequences, tests, and remedies
• properties of nonstationary time series
• tests for nonstationarity
• cointegration
• error correction models

The course will conclude with a brief treatment of panel data regressions.
The course will not use linear (matrix) algebra. 

This course will consist of three hours of lectures each morning and a small-group class lasting one hour and a half each afternoon. A written problem set will be due for each class.

A 2-hour examination will take place on the afternoon of Friday 28th August 2015.

This course is offered as part of the LSE Methods Summer Programme, a summer school of intensive short courses in social science research methods for students, researchers and professionals. A number of social events will be held throughout the programme. Participants will be provided with a transcript and certificate upon completion of the course.

Cost:

Students: £1,435
Academic/charity staff: £1,930
Professionals: £2,425

Website and registration:

http://www.lse.ac.uk/study/summerSchools/Methods/Economists/ME107.aspx

Region:

Greater London

Keywords:

Linear regression, Probit regression, Panel data models, Time Series Analysis, Econometrics, SPSS

Related publications and presentations:

Linear regression
Probit regression
Panel data models
Time Series Analysis
Econometrics
SPSS

Back to archive...