Econometric Theory
As Taught Autumn
Semester 2011
Professor S Leybourne, School of Economics, University of Nottingham
Module
Title: Econometric Theory
Module
Code: L14003
Total
Credits: 15
Level
of Study: Post Graduate
Offering
School: School of
Economics
Frequency
of Class: 1 x 2 hour
lectures per week, 1 x 1 hour tutorial per week
Targeted Students: MSc students in the School of
Economics.
The content
presented here provides information for prospective students on module L14003 –
‘Econometric Theory’, offered by the School of Economics, University of
Nottingham. The module convenor is Professor S Leybourne.
This module
generalises and builds upon the econometric techniques applied to the
multivariate regression models covered in undergraduate econometric courses, by
introducing a number of new statistical and econometric concepts. Following a
review of the classical linear regression model, students will cover
alternative estimation methods (Maximum Likelihood estimation), some basic
elements of asymptotic theory (the weak law of large numbers and the Central
Limit Theorem), and the consistency and asymptotic normality of OLS. In the
second part of the module, attention will switch to introductory econometric
methods applicable to economic time series - for example, monthly or quarterly
economic data. In particular, the model building approach for a single time
series, based on the methodology of fitting and estimating ARMA models, will be
introduced, and its application to forecasting discussed.
The
following topics will be covered in the module:
Classical
Linear Regression Model
Classical assumptions of the model; Ordinary least squares (OLS) estimation;
Gauss-Markov Theorem; Hypothesis testing, Maximum likelihood (ML) estimation;
Restricted OLS estimation.
Asymptotic
Theory
Modes of convergence; The weak law of large numbers; The central limit theorem;
Consistency and asymptotic normality of OLS.
Autogressive
Moving Average (ARMA) Modelling.
Stochastic processes; AR, MA and ARMA models; Stationarity and Invertibility;
Autocorrelation functions; Yule-Walker equations; Nonstationary processes.
Identification
and Estimation of ARMA Models.
Sample and partial autocorrelation functions; Least squares estimation;
prediction error decomposition; Conditional likelihood function; Maximum
likelihood estimation.
Assessing
Model Fit.
Tests for randomness of residuals; Information criteria.
Forecasting.
Optimal forecasts; Forecast functions.