Content and objectives
The challenge of econometrics is to find out what everyday reality, properly recorded and interpreted, can tell us about the relevance of economic and social theories. Governance theories concern, more often than not, the effects of economic and social policies. It is the purpose of econometric methods to measure the impact and assess the effectiveness of policy interventions.

The snag of econometric methods is that they depend on statistical induction. Apart from being fraught with technicalities, statistical induction is unavoidably subject to statistical error. Whereas a mathematical conclusion is arguably “exact”, a statistical conclusion is not – even though deep mathematical arguments may have been invoked in the process. Thus every decent statistical estimate comes with a “standard error” attached to it, revealing the scale of the sampling error it contains. Furthermore, ordinary sampling error is a trifle as compared to some other kinds of statistical error, called (imaginatively) non-sampling errors. These are much harder to control and require more subtle methods of statistical analysis. Unfortunately non-sampling errors are conspicuously present in economic and social policy research.

The subject matter of this course is the use of (mainstream) econometric methods in economic and social policy research. Some prior exposure to econometrics at least at an introductory level is assumed. This should include the mathematical formulation of economic theories as well as the concept of a statistical model. The course explicitly aims at combining theoretical insight with empirical practice and group activity. Participants will study the theory, read about applications, deal with actual data using econometric software (Stata or EViews), prepare a presentation and conduct discussions. Hence a lot of personal involvement and social commitment will be demanded. A willingness to think formally and make one’s hands dirty is essential.

The course comprises (in principle) a series of 12 meetings spread out over the months of November and December. About two thirds of the meeting time will be taken up by lectures and reviews of classical econometric theory and methods. In between meetings participants will be required to study the theory but also to read about various applications and to do actual data work using econometric software (Stata or Eviews). In the remaining one third of the meeting time participants will briefly present one application themselves (possibly even a research paper of their own) and take an active part in seminar discussions. The topic of their application may be selected from the list of articles provided below (which can be updated at any time).

Minimum Requirements

• Prerequisites: Basic algebra, calculus, basic probability and statistics, and classical regression analysis. It will certainly help if you also had some training in matrix algebra, mathematical statistics, and general linear models. A willingness to think formally as well as to make one’s hands dirty is essential.

• All participants are explicitly required to read the first four(introductory) chapters of Stock & Watson (2007) before the beginning of the course. In the first meeting we will focus directly on the regression model (starting from Chapter 4)

• Attendance, active participation and commitment, assignments, presentations and discussions, as well as a final paper, are all part of the course requirements. Students with an advanced background in theoretical and empirical econometrics may be allowed to substitute a research paper for the assignments but are requested to participate in the presentations and discussions anyway.

Literature

• François Bourguignon, Francisco H. G. Ferreira and Phillippe G. Leite, 2003, “Conditional Cash Transfers, Schooling and Child Labor: Micro-Simulating Bolsa Escola”, DELTA Working Paper 2003-07

• Bourguignon, François, and Luiz A. Pereira da Silva. 2003. The Impact of Economic Policies on Poverty and Income Distribution: Evaluation Techniques and Tools. World Bank and Oxford University Press, Washington, D.C. and New York. http://www1.worldbank.org/publications/pdfs/15491frontmat.pdf

• Cockburn, J. Trade liberalisation and poverty in Nepal: A Computable General Equilibrium Micro Analysis. Discussion paper 01-18, Centre de Recherche en  Économie et Finance Appliquées (Université Laval), October 2001. http://www.crefa.ecn.ulaval.ca/cahier/0118.pdf 

• Davies, Jim, Microsimulation, CGE and Macro Modelling for Transition and Developing Economies, mimeo, University of Western Ontario, 2004.

• O’Donoghue, C., 2005. Social Policy Modelling Lecture Notes. Circulated in Advance of Lectures