Econometric Analysis and Applications

Econometric Analysis and Applications is the second econometrics module offered to MSc students who need to broaden their understanding of the application of quantitative methods to inquiry in finance or economics. This module assumes that you have studied the classical linear regression model at an introductory level and that you are familiar with the assumptions that underlie that model. You will be aware that there are many cases in which these assumptions are not satisfied, and know how such problems as heteroscedastic disturbances and autocorrelated errors can be detected, and what can be done about them.

The purpose of this module is to broaden your knowledge and extend your understanding of econometrics. In doing this, you will work with data. The first two units extend your knowledge of single equation methods. Unit 1 considers how to make progress with dummy – that is, qualitative – regressors. Unit 2 introduces dynamic models by showing how lags and expectations can be incorporated. The following three units focus on models that consist of two or more equations – simultaneous equation models. The nature of such systems is explained and their identification and estimation discussed. The analysis of dynamic models is extended in Unit 6, where the time series properties of variables are discussed. By understanding the time series properties of variables, you will be able to specify dynamic econometric models that capture both short-run and long-run effects. These are discussed in Unit 7. The final unit, Unit 8, focuses on forecasting with both econometric and time series models.

We recommend that you study Econometric Principles and Data Analysis prior to this module.

Learning outcomes

After studying this module you will be able to:

• explain the use of intercept and slope dummy variables
• use and interpret the Chow test of parameter stability
• explain the nature of the ‘dummy variable trap’ and how to avoid it
• explain finite distributed lag models, including immediate impact, long-run reactions and mean lag
• discuss the properties of estimators of distributed lag and autoregressive models
• implement both Durbin’s h test and the LM test of autocorrelation and interpret the results
• explain ‘simultaneous equation bias’
• interpret in a model the behavioural equations, definitions or identities, and equilibrium conditions
• examine the structural form, reduced form and final form of a simultaneous equation system
• identify conditions for stability in dynamic simultaneous equation systems
• explain the identification problem
• discuss the implications of equations which are exactly identified, overidentified, and not identified
• explain and apply indirect least squares
• explain the method of instrumental variable estimation, and discuss the properties of IV estimators
• explain and apply the method of two stage least squares (TSLS or 2SLS), and discuss the properties of 2SLS estimators
• discuss what is meant by stationary and nonstationary time series, and provide examples of each
• explain and implement the Dickey–Fuller and augmented Dickey–Fuller tests of the hypothesis that a series is I(1)
• explain the nature of cointegration and the relationship between spurious regression and cointegration
• discuss and implement tests of cointegration
• explain, interpret and estimate error correction models
• explain the nature of vector autoregressions (VARs)
• define an autoregressive integrated moving average (ARIMA) model, and use it to forecast
• interpret measures of forecast evaluation.

Tuition and assessment

Students are individually assigned an academic tutor for the duration of the module, with whom you can discuss academic queries at regular intervals during the study session.

You are required to complete two Assignments for this module, which will be marked by your tutor. Assignments are each worth 15% of your total mark. You will be expected to submit your first assignment by the Tuesday of Week 6, and the second assignment at the end of the module, on the Tuesday after Week 10. Assignments are submitted and feedback given online. In addition, queries and problems can be answered through the Virtual Learning Environment.

You will also sit a three-hour examination on a specified date in September/October, worth 70% of your total mark. An up-to-date timetable of examinations is published on the website in July each year.

Study resources

• Study guide: The module study guide is carefully structured to provide the main teaching, defining and exploring the main concepts and issues, locating these within current debate and introducing and linking the assigned readings.
• Key text: Jeffrey M Wooldridge (2020) Introductory Econometrics. 7th Edition. Boston MA: Cengage.
• Software: This module will use R. This is a widely used programming environment for data analysis and graphics. You will use this software to do the exercises in the units, and also the data analysis part of your assignments. The results presented in the units are also from R.
• Exercises: There are exercises in every unit. Many of these exercises require you to work with R and data files, available from the VLE, to do your own econometric analysis.
• Virtual learning environment: You will have access to the VLE, a web-accessed study centre. Via the VLE, you can communicate with your assigned academic tutor, administrators and other students on the module using discussion forums. The VLE also provides access to the module Study Guide and assignments, as well as a selection of electronic journals available on the University of London Online Library.

Study calendar 2022/23

Study calendars are subject to change.

Module overview

Unit 1 Dummy Variables

• 1.1 Ideas and Issues
• 1.2 The Use of Dummy Variables
• 1.3 The Chow Test for Parameter Stability
• 1.4 Unit Study Guide
• 1.5 Example – Long-Term Trends in Terms of Trade
• 1.6 Conclusion
• 1.7 Exercises
• 1.8 Answers to Exercises

Unit 2 Dynamic Models – Lags and Expectations

• 2.1 Ideas and Issues
• 2.2 Lags
• 2.3 Expectations
• 2.4 Properties of OLS Estimators
• 2.5 Causality – The Granger Test
• 2.6 Unit Study Guide
• 2.7 Example – Long-Term and Short-Term Interest Rates
• 2.8 Conclusion
• 2.9 Exercises
• 2.10 Answers to Exercises

Unit 3 Simultaneous Equation Models

• 3.1 Ideas and Issues
• 3.2 Unit Study Guide
• 3.3 Example – The Polak Model
• 3.4 Conclusion
• 3.5 Exercises
• 3.6 Answers to Exercises

Unit 4 The Identification Problem

• 4.1 Ideas and Issues
• 4.2 Unit Study Guide
• 4.3 Example – Bid–Ask Spreads and Trading Activity in Options
• 4.4 Conclusion
• 4.5 Exercises
• 4.6 Answers to Exercises
• Appendix – Estimates of Moore’s 1914 Model

Unit 5 Simultaneous Equation Models – Estimation

• 5.1 Ideas and Issues
• 5.2 Unit Study Guide
• 5.3 Example – Stock Market Returns and Bond Yields
• 5.4 Conclusion
• 5.5 Exercises
• 5.6 Answers to Exercises

Unit 6 Univariate Time Series – Stationarity and Nonstationarity

• 6.1 Ideas and Issues
• 6.2 Stationary and Nonstationary Time Series
• 6.3 Integrated and Trend Stationary Series
• 6.4 The Nature of Financial Data
• 6.5 Correlograms
• 6.6 Unit Root Tests
• 6.7 Examples – Simulated Series
• 6.8 Example – Exchange rate US$ to UK£
• 6.9 A Procedure for Unit Root Tests
• 6.10 Conclusion
• 6.11 Exercises
• 6.12 Answers to Exercises

Unit 7 Multivariate Time Series Analysis

• 7.1 Ideas and Issues
• 7.2 The Engle–Granger Approach
• 7.3 Error Correction Models
• 7.4 The Johansen Approach
• 7.5 Example – The Single Index Model
• 7.6 Example – UK Financial Markets
• 7.7 Conclusion
• 7.8 Exercises
• 7.9 Answers to Exercises
• Appendix – 0.05 Critical Values: Unit Root and Engle–Granger Cointegration Tests

Unit 8 Forecasting

• 8.1 Ideas and Issues
• 8.2 Example – Forecasting Earnings
• 8.3 Conclusion
• 8.4 Exercises
• 8.5 Answers to Exercises

Disclaimer

Important notice regarding changes to programmes and modules