Introduction to Econometrics

1
Introduction to Econometrics

• Lecture 1
• Introduction and overview of the course
• Definition, scope and methodology of
  econometrics
• A review of the simple (bivariate) linear
  regression model

2
Objectives

• To provide you with information about
• the subject of econometrics and the topics that
  we shall cover in the unit
• the learning teaching and assessment
  arrangements for the unit
• To review the simple bivariate linear
• regression model

3
Learning and Teaching

• Lectures and accompanying notes
• Problems classes (seminars)
• Computer lab sessions
• Office hours and econometrics café sessions
• Text books
• Web pages and links
• the econmet wiki
• One minute e-mail and FAQs

4
Recommended texts

• Dougherty, C (2007) Introduction to Econometrics,
  Third Edition, OUP
• Gujarati, D N and Porter, DC (2009) Basic
  Econometrics, Fifth Edition, McGraw-Hill
• Kennedy, P (2003) A Guide to Econometrics, Fifth
  Edition, Blackwell
• Koop, G (2008) Introduction to Econometrics, John
  Wiley Sons
• and others available in the Library in section
  330.0182

5
Computer software for the lab classes

• The regression tool in Excel
• PcGive
• EViews

6
assessment

• end of unit exam (50 weighting)
• portfolio of practical solutions (20 weighting)
• assignment/report (30 weighting)

7
econometrics

• The measurement of economic relationships

8
econometrics

• the application of mathematical statistics to
  economic data to lend empirical support to models
  constructed by mathematical economics and to
  obtain numerical estimates (Samuelson et al.,
  Econometrica, 1954)

9
aims of econometric modelling

• explanation
• policy evaluation
• forecasting

10
types of data

• cross-section
• time-series
• panel

11
types of model

• simple bivariate linear
• non-linear bivariate
• multiple regression
• dynamic
• simultaneous equation
• other (e.g. logit and probit)
• panel

12
Example sales-advertising relationship

• suppose we wish to test the hypothesis that a
  firms sales are dependent upon its advertising
• the simplest model is
• sales a badvertising u
• where a and b are parameters to be estimated, u
  is an unobservable error term - a random
  disturbance
• this is an example of a simple bivariate
  regression model

13
sales and advertising time series plot
14
sales and advertising scatter diagram
15
general notation for the simple bivariate linear
model

for i 1,2,.n
With time series data we tend use t rather than i
as the subscript and T as the sample size
16
model specification

• the equation(s) variables and functional form
• a priori restrictions on parameters
• stochastic assumptions (assumptions about the
  disturbance term)

17
assumptions about u

• mean zero
• constant variance
• independent between observations
• independent of the X variable

18
The role of the disturbance term

• Reasons for the disturbance
• omitted influences on Y
• errors of measurement
• errors in variables
• non-linearity
• random nature of human behaviour

19
Econometric problems

• Reasons for the disturbance
• autocorrelation
• heteroskedasticity
• bias
• multicollinearity

20
The unlikely case of control over sample design
Plant X Y FITTED Y 1 100 40 42.32 2
200 50 48.21 3 300 50 54.11 4 400 70
60.00 5 500 65 65.89 6 600 65 71.79
7 700 80 77.68