Econometric Analysis

1
Econometric Analysis
Lecture 1 Introduction and overview
2
Objectives

• To inform you about the aims and scope of the
  unit
• To explain the learning, teaching and assessment
  arrangements
• To review some of the basics of econometrics
  covered in the Introduction to Econometrics unit
• To review the use of PcGive for practical
  econometrics work with an interactive example

3
Learning and teaching

• Lectures and accompanying notes
• Computer lab sessions
• Surgery and office hours
• Text books
• Web pages and links, Victory site and the L drive
• One minute e-mail and FAQs

4
Provisional Lecture Programme
Week Topic 1 Introduction and overview 2 A
review of single equation regression models for
cross-section data 3 Further aspects of
cross-section data econometric models 4
Regression models with time-series data 5 Further
aspects of time-series data econometric models 6
Testing for unit roots, Granger causality and
cointegration 7 Instrumental variables and
simultaneous equation models 8 More on
Instrumental variables and simultaneous equation
models 9 Limited dependent variables models 10
Pooled and panel data models 11 Review and
Revision
5
Recommended texts

• Main text
• Wooldridge, J M (2009) Introductory Econometrics.
  A Modern Approach, South-Western CENGAGE Learning
  4th edition (or alternatively the 3rd edition,
  2006)
• Other useful books include
• Dougherty, C (2007) Introduction to Econometrics,
  Third Edition, OUP
• Gujarati, D N (2002) Basic Econometrics, Fourth
  Edition, McGraw-Hill
• Kennedy, P (2003) A Guide to Econometrics, Fifth
  Edition, Blackwell
• And others available in the Library in section
  330.0182

6
Assessment

• end of unit exam (50 weighting)
• two reports (25 weighting each) providing
  extended analysis of some of the exercises
  covered in the weekly practical classes

7
Section 1 (weeks 1-3) Wooldridge chapters 1-9
econometric models based on cross-section data

• a review of single equation econometric models
  for cross-section data
• standard classical assumptions relating to
  independence of disturbances, homoskedasticity,
  normality etc.
• discussion of data issues quantitative and
  qualitative variables proxies and dummies
  dealing with missing data and outliers units of
  measurement nominal, real, per capita data etc
  multicollinearity and micronumerosity issues -
  adjustment of data to create an X matrix with
  suitable measurement properties
• model formulation issues functional form,
  transformations, a priori parameter assumptions
  and possible restrictions (zero or exclusion
  restrictions, other linear restrictions)
• tests of restrictions (specification tests) and
  diagnostic testing (misspecification tests)
  Chow tests
• the general to specific approach to modelling
  a process for arriving at a model that is
  congruent (theory-consistent, data admissible,
  data coherent, encompassing and has predictive
  validity)

8
Interactive illustration modelling variations
in wages in a cross-section data set

• The data set wage1 comprises observations on 24
  variables for 540 individuals in the US in 1976,
  drawn from the Current Population Survey (some of
  these variables are just transformations of other
  variables in the data set)
• The data are available in two formats Excel
  (wage1.xls) and the special PcGive format
  (wage1.in7)
• Variable definitions may be found in the file
  wage1.des
• We will use this file to discuss points on the
  previous slide and to illustrate the use of
  PcGive.