Centre for Financial & Management Studies (CeFiMS) - University of London

Individual Professional Courses – IPC  

Econometric Analysis & Applications [FE204]

Introduction

Econometric Analysis and Applications is the second econometrics course that can be taken as part of the MSc in Financial Economics. It extends the basic introduction to econometric analysis developed in the core course, Econometric Principles and Data Analysis. This course teaches the more advanced techniques of dummy variables, lags and expectations, simultaneous equation models, non-stationarity and co-integration and forecasting. The course ends with a brief discussion of ‘further topics for econometrics’ for students who are particularly keen to develop their quantitative skills beyond the course. It assumes that you have studied the classical linear regression model at an introductory level and that you are familiar with the assumptions which underlie the model. It is also assumed that you have a basic working knowledge of the econometric software, MICROFIT. There are many examples to illustrate the main themes in a way which will help you in both understanding the econometrics and putting the theory to use with data.

Aims & Objectives

This course aims to broaden your knowledge and extend your understanding of econometrics.

By the end of the course you should be able to:

• make progress with qualitative regressors, dummy variables and the identification and estimation of simultaneous econometric models
• show how lags and expectations can be incorporated in dynamic models
• forecast with both econometric and time series models.

Resources

Students receive a looseleaf binder containing eight ‘course units’; these texts are carefully structured to provide the main teaching and are equivalent to traditional course lectures, defining and exploring the main concepts and issues, locating these within current economics debate and introducing and linking the further assigned readings. Two obligatory assignments, which are marked by your CeFiMS tutors, and a specimen examination paper are also sent to students, along with the following:

Textbook:

Damodar N. Gujarati, Basic Econometrics, third edition, McGraw-Hill International, 1996, ISBN0071139648.

Computer Programme:

The software used to carry out econometric exercises for this course is MICROFIT, a very user-friendly package, for which full guidance is provided. To run this program, you need access to a computer which is IBM (PC) compatible, with a hard disk drive (and Microsoft Windows 95 or higher operating system).

Course Timetable:

This shows the linkage between the various components of the course.

Course Content

Unit 1 Dummy Variables

This unit shows how explanatory variables which are qualitative, rather than quantitative in nature (such as ethnic group, sex or war) can be included in regression analysis. This is done by the use of dummy variables which are qualitative or categorical in nature. It shows how to avoid the ‘dummy variable trap’, and the use of the Chow test of parameter instability is also introduced in this unit.

Unit 2 Dynamic Models

Lags and Expectations

Unit 2 introduces dynamic models by showing how lags and expectations can be incorporated in linear regression. The unit aims to introduce students to finite and infinite distributed lag models; the Koyck transformation; partial adjustment; error correction models the adaptive expectations hypothesis; how partial adjustment and adaptive expectations hypotheses provide a rationale for the Koyck transformation; direct measures of expectations; the properties of estimators of distributed lag and autoregressive models; the estimation of lagged models and how to interpret the results; the implementation of Durbin’s h test, the LM test of autocorrelation, and the Granger test of causality and the interpretation of their results.

Unit 3 Simultaneous Equation Models

Unit 3 focuses on models which consist of two or more equations: simultaneous equation models. The unit also discusses the distinction that can be drawn between static and dynamic models, including lagged endogenous variables.

Unit 4 The Identification Problem

Unit 4 introduces the concept of identification. The precise conditions for identification are explained using the order and rank conditions. The question of identification is related to the problem of estimating the parameters of a structural equation.

Unit 5 Simultaneous Equation Models

Estimation

Unit 5 introduces limited information (single equation) methods for estimating an exactly identified or overidentified equation in a simultaneous model. Three frequently used methods are discussed: OLS (ordinary least squares), ILS (indirect least squares), 2SLS (two stage least squares). The unit teaches the properties of the OLS estimator of the slope coefficients of a structural equation from a simultaneous system; how to describe a recursive model and explain why it can be estimated by OLS; how to explain the method of ILS, implement it in appropriate situations and understand the properties of ILS estimators; it explains and discusses the method of 2SLS, how to implement it and understand the properties of 2SLS estimators; and how to interpret Sargan’s chi square test of model specification.

Unit 6 Univariate Time Series

Stationarity and Non-stationarity

Unit 6 discusses the time series properties of variables. The unit aims to teach an understanding of: the difference between stationary and nonstationary series; the nature of integrated series, in particular I(1) and I(0) series; the difference between I(1) series and trend stationary series; how to test whether a series is I(1) or trend stationary; the concept of cointegration between I(1) series; and formal tests of nonstationarity. The unit covers Dickey-Fuller , augmented Dickey-fuller and Phillips-Perron unit root tests.

Unit 7 Multivariate Time Series Analysis

Unit 7 focuses on cointegration and error correction model (ECMs). It includes tests of cointegration, interpretation of ECMs, four methods of estimation for ECMs, and their limitations; explanation of VARs, and an exposition of Johansen cointegration tests and the limitations of the Johansen approach.

Unit 8 Forecasting

Unit 8 uses time series and econometric models to generate forecasts. It explores the difference between time series and econometric models; discusses the circumstances in which time series and econometric models are used; explains what autoregressive process and moving average processes are; teaches what an ARIMA model is and how to use it to forecast; discusses the sources of error in a forecast generated from an econometric model; teaches students to interpret the measures of forecast evaluation that MICROFIT calculates; explains the relationship between a model in structural, restricted and unrestricted reduced form; shows how to estimate AR(1) and MA(1) models and use them for forecasting; demonstrates how to calculate forecasts using econometric models, and how to understand static and dynamic single equation and simultaneous systems.

Tuition & Assessment

The course is assessed by two assignments and a three-hour examination held in the autumn. Assignments count for 30%, while the examination accounts for 70% of the final mark.