- Department: Economics and Related Studies
- Module co-ordinator: Dr. Vanessa Smith
- Credit value: 20 credits
- Credit level: H
- Academic year of delivery: 2023-24
- See module specification for other years: 2024-25
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.
|Semester 2 2023-24
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.
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
Outline of content
Instrumental variable methods when regressors are endogenous
Binary choice models (Logit and Probit) for modeling probabilities
Stationary time series models: AR, MA and ARMA models; forecasting and evaluation of forecast performance
Non-stationary time series and testing for unit roots
Cointegration analysis in the case of a single equation model (Engle-Granger approach and ARDL)
Volatility modeling using GARCH models
Panel data models
|% of module mark
Formative assessment will consist of two submitted seminar exercises
|% of module mark
Individual feedback will be given on the project as marks are released, in line with DERS policy, as well as on their formative assessment.
Stock, J H. & Watson, M W. (2015). Introduction to Econometrics (3rd ed.), Pearson International ed.
Enders, W. (2015). Applied Econometric Time Series (4th ed.). Wiley
Verbeek. A Guide to Modern Econometrics