402762 Angewandte konometrie Introduction to Applied Econometrics

1
40276/2 Angewandte Ökonometrie Introduction to
Applied Econometrics 

• Dr. Julia Wörzwiiw Wiener Institut für
  Internationale WirtschaftsvergleicheOppolzergasse
  6, A-1010 WienTel 533 66 10 24E-mail
  woerz_at_wiiw.ac.at Sprechstunde nach Vereinbarung
   Time Place Friday, 1300-1500, SR1
  Hohenstaufengasse

2
Description
The course introduces the main workhorse of
applied empirical research in economics, linear
regression by OLS Ordinary least Squares.
After having taken the course, students should
understand and be able to evaluate applied
analysis of cross-section data and be able to
understand such analysis themselves.  




.

3
Assessment
2 take-home exercises - each 25 of final
grade 1 group project - 50 of final grade -gt
to be handed in in a written form and to be
presented at the end of the term. A minimum of
51 is needed for a positive evaluation!    




.

4
Text Book

• Stock James H. and Mark W. Watson Introduction
  to Econometrics (Addison Wesley)
• Greene, William H. Econometric Analysis
  (McGraw Hill)
• Johnston, Jack and John DiNardo Econometric
  Methods (McGraw Hill)
• Verbeek, Marno A guide to Modern Econometrics
  (Wiley)
• Ben Vogelvang Econometrics Theory and
  Applications with EViews (Pearson Education
  Ltd.)
• Hackl Peter Einführung in die Ökonometrie
  (Pearson Studium München)

Additional Reading
.

5
Course Outline

• Revision Statistics and Probability Theory
• Single Variable Regression Model Ordinary Least
  Squares, Gauss-Markov-Theorem
• Hypothesis Testing, Confidence Intervals
• Introduction into EVIEWS
• Binary regressors, heteroscedasticity, omitted
  variable bias
• Multiple Variable Regression Model, more
  problems multicollinearity
• Non-linear regression functions, interaction
  terms
• Diagnostic checking
• Instrumental Variable Estimation (short
  introduction)

.

6
Schedule

• 11 units (?) check April 13, June 8
• No class on
• ? March 30 2007
• ? June 29 2007 
• Group work Data sets on
• economic growth
• financial development (macro data)
• CEECs
• exchange rates
• US labour market

.

7
Introduction
David Hendry Econometrics Alchemy or
Science? Oxford University Press,
2000. Econometrics ? econo-mystics ?
economic-tricks Defintion ...the advancement
of economic theory in its relation to statistics
and mathematics. (Econometric Society, founded
1930)




.

8
Comment by Keynes (1940) on Tinbergens book
(1939)
No one could be more frank, more painstaking,
more free from subjective bias or parti pris than
Professor Tinbergen. There is no one, therefore,
as far as human qualities go, whom it would be
safer to trust with black magic. That there is
anyone I would trust with it at the present
stage, or that this brand of statistical alchemy
is ripe to become a branch of science, I am not
yet persuaded. But Newton, Boyle and Locke all
played with Alchemy. So let him continue.




.

9
Econometrics as Alchemy
Philosphers Stone of the econometrician REGRES
SION ANALYSIS -gt transforms data into
significant results




.

10
Hendrys Theory of Inflation in the UK
C causes inflation (P)




P...consumers expenditure deflator
.

11
Hendrys Theory of Inflation in the UK
Fit a model P 10.9 3.2 C 0.39 C2
R2 0.982 (0.55) (0.23)
(0.02) d 0.1

• C is highly significant
• Good fit
• Problem autocorrelation!

.

12
Hendrys Theory of Inflation in the UK
Fit 1st order AR process Pt 21.1 7.2 Ct
0.78 Ct2 0.94Ut-1 R2
0.998 (3.6) (1.4) (0.13)
(0.02) d 1.9

• C is highly significant
• Spectacular fit, no AR left, BUT ...

.

13
... BUT
C is simply cumulative rainfall in the UK  ?




.

14
Back to Keynes critique

• problems of linear regression model
• Using an incomplete set of determining variables
  ...
• Building models with unobservable variables ...
• Estimated from badly measured data ...
• Obtaining spurios correlations ...
• Unable to separate ... multicollinear variables
  ...
• Assuming functional forms ... not appropriate ...
• Misspecifying dynamic reactions ...
• Invalidly inferring causes from correlations
  ...
• Confusing statistical with economic
  significance

.

15
Still need more motivation for this course?
?
That the subject is exceedingly complicated does
not entail it is hopeless. (Hendry, 2000)




.

16
economic theory statistics / data statistical
methods interpretation   ?   econometrics




.

17
Types of Data




.

18
Types of Data




.

19
Types of data  Time series time
dimension Cross section units, sampling! Panel
data unit time dimension   Types
of econometric models  Time series models
Behavioural models Cross-section models
Panel data models




.