Introductory Econometrics


Econometrics aims to integrate economic theory, statistics, some mathematics and real-world data. For example, from economic theory we deduce (1) an inverse relationship between price and the quantity demanded; (2) when the interest rate decreases, planned investment expenditure increases, and (3) that consumption expenditure is positively related to income, etc.

However, theory generally does not tell us by how much these variables change in reaction to a change in one of these determining factors. Econometrics is an extremely useful tool, in that it allows us to estimate the magnitude and strength of the relationships that exist between various sets of variables, and/or whether the relationship exists at all.

Econometrics has much practical application, not only as an academic endeavour, but also in the government and private sectors. Your potential new employer would not be particularly impressed if you tell him that total sales revenue will increase if you lower the price of the product. He would be far more impressed if you tell him that sales revenue is likely to increase by 6 per cent for every 10 per cent decrease in the price. Another example from the finance/asset management field: If you believe that the rand is going to depreciate rapidly, you would want to be in "rand-hedge" shares. Typically these would be resource shares. Econometrically it is possible to determine, based on historical data, which shares have the greatest sensitivity to a change in the exchange rate.

The aim of this course is to give you a solid introduction to applied econometrics. The focus will not primarily be on the statistical derivations, but rather on the applications of the econometric theory. Computer tutorials and projects comprise a significant component of the course.

Econometrics as we know it can trace its origins to the late 1930s. However, during the past two decades there have been major developments, both of a practical and theoretical nature. The advances in computer hardware and software during this period have resulted in the expansion of techniques that would not have been dreamed of twenty or even ten years ago. As a result of these developments, traditional econometrics has come under fire.

Specifically, the fact that traditional econometrics generally does not investigate whether the data are stationary (i.e. the mean and standard deviation do not vary over time) has drawn much criticism in recent years. New approaches, specifically those of cointegration, have been developed to deal with this problem.

Despite the problems associated with traditional econometrics, there does not seem to be a better way to teach the basic principles of econometrics in any other way. The first three-quarters of this course will focus on basic and generic econometric skills. The last quarter of the course will focus on some more recent developments in econometrics.

Course instructors: Dieter von Fintel and Willem Boshoff

Work programme

BER Weekly

18 February 2019
There was a slew of data releases last week, both on the domestic and international front. Domestic releases focused on real economic data for December 2018, all but completing the picture for 2018Q4 GDP. In financial markets, the rand lost further ground last week on the back of a strong US dollar and the return of load shedding. On the international...

Read the full issue