Applied Econometrics - ECO00040H

«Back to module search

Module summary

An introduction to the key statistical techniques required to examine economic models with data, enabling students to follow large parts of the empirical literature and to carry out such analyses themselves.

Related modules

Module will run

Occurrence	Teaching period
A	Semester 2 2025-26

Module aims

To develop skills needed to apply econometric techniques in the following contexts: (i) the implementation of instrumental variable methods when regressors are endogenous; (ii) the use of binary choice models to model probabilities in applied economics; (iii) the estimation and interpretation of models designed for panel data; (iv) forecasting using stationary ARMA models and evaluating forecast performance; (v) the investigation of the time series properties of economic data and the implications of these properties for least squares analysis; and (vi) cointegration analyses when a single equation model is under scrutiny and the derivation of associated error correction schemes when variables are cointegrated.

To develop skills needed to interpret applied econometric results in the following contexts: (i) the analysis of regression models in the presence of omitted variables; (ii) the application of linear probability, logit and probit models; (iii) relationships estimated using linear panel data models; (iv) the outcomes of a battery of diagnostic checks after estimation; (v) testing for unit roots in economic variables by means of Dickey-Fuller tests; and (vi) empirical analyses based upon either the Granger-Engle two-step method or the Autoregressive Distributed Lag model.

Module learning outcomes

On completing the module a student will be able to:

• Read and understand more of the econometric evidence published in academic journals and books. Understanding is extended beyond the second year Econometrics for Economists module by covering new topics such as: instrumental variable methods; binary choice models; and panel data (in which there are both cross-section and time series dimensions); forecasting using stationary dynamic ARMA models and evaluating forecast performance; nonstationary time series variables in regression; integration and cointegration (which are very important in modern applied macroeconomics).

• Use standard econometric software (seminar work will involve the use of popular econometrics packages with various data sets that are provided via links on the VLE page)

• Formulate economic hypotheses in testable ways and to understand which methods are appropriate for carrying out statistical tests

Module content

Outline of content

1. Instrumental variable methods when regressors are endogenous

2. Binary choice models (Logit and Probit) for modeling probabilities

3. Stationary time series models: AR, MA and ARMA models; forecasting and evaluation of forecast performance

4. Non-stationary time series and testing for unit roots

5. Cointegration analysis in the case of a single equation model (Engle-Granger approach and ARDL)

6. Volatility modeling using GARCH models

7. Panel data models

Indicative assessment

Task	% of module mark
Essay/coursework	100

Special assessment rules

None

Additional assessment information

Formative assessment will consist of two submitted seminar exercises

Indicative reassessment

Task	% of module mark
Essay/coursework	100

Module feedback

Individual feedback will be given on the project as marks are released, in line with DERS policy, as well as on their formative assessment.

Indicative reading

1. Stock, J H. & Watson, M W. (2015). Introduction to Econometrics (3rd ed.), Pearson International ed.

2. Enders, W. (2015). Applied Econometric Time Series (4th ed.). Wiley

3. Verbeek. A Guide to Modern Econometrics