Source document. This page reproduces the syllabus this course was written to, as published — its semesters, credits and paper numbers are that document’s, not this site’s. The course itself is studied on its own, in any order.
Course Overview
Title
Econometrics
Semester
VIII (additional / advanced course beyond the core syllabus)
Theory Credits
4 (5 hrs/week)
Pre-requisites
Statistical Methods, Inferential Statistics, Continuous Distributions, ideally R Programming
Program Objectives
To introduce the discipline of econometrics — the intersection of economic theory, mathematics, and statistics.
To equip students with the ability to specify, estimate, test and interpret single-equation linear econometric models.
To develop diagnostic skills for heteroscedasticity, multicollinearity and autocorrelation, and their remedies.
Learning Outcomes
After successful completion of this course, students will be able to:
Identify the four data structures used in econometrics (cross-section, pooled cross-section, time-series, panel) and recognise the issues each brings.
Estimate two-variable and multiple linear regression models by OLS, derive the Gauss–Markov theorem, and demonstrate the BLUE property.
Use \(R^2\), \(t\)- and \(F\)-tests to evaluate model adequacy; understand the ANOVA decomposition.
Diagnose heteroscedasticity (Park, Glejser, Breusch–Pagan, White, Goldfeld–Quandt) and apply WLS or robust SEs.
Measure and remedy multicollinearity using VIF, tolerance, ridge regression, principal components.
Test for autocorrelation (Durbin–Watson, Breusch–Godfrey), estimate the AR(1) coefficient, and apply Cochrane–Orcutt or HAC SEs.
Theory — Five Units
Unit I: Basic Econometrics
Nature of econometrics and economic data; concept of econometrics; steps in an empirical economic analysis; the econometric model; importance of measurement in economics; structure of econometric data — cross-section, pooled cross-section, time-series, paired (panel) data.
Simple regression models — two-variable linear regression model, assumptions and estimation of parameters; Gauss–Markov theorem; OLS estimations; partial and multiple correlation coefficients; the general linear model — assumptions, estimation and properties of estimators; BLUEs.
Tests of significance of estimators; \(R^2\) and ANOVA; concepts and consequences of heteroscedasticity; tests (Park, Glejser, Breusch–Pagan, White, Goldfeld–Quandt) and solutions (WLS, robust SEs); specification error; errors of measurement.
Concept of multicollinearity and its consequences on econometric models; detection methods; variance inflation factor (VIF) and tolerance — formula and interpretation; methods of reducing the influence of multicollinearity (dropping variables, ridge regression, principal components, transformation).
Disturbance term in economic models and its assumptions; consequences of autocorrelated disturbances; detecting autocorrelation — hypothesis tests; Durbin–Watson test; estimation of autocorrelation coefficient for a first-order autoregressive scheme; Breusch–Godfrey LM test; Cochrane–Orcutt iterative procedure; Newey–West HAC standard errors.