Introduction to Econometrics

Undergraduate Program
Department of Economics
Universitas Padjadjaran
 Instructors
• Your instructor
– Ekki Syamsulhakim
• [email protected]
• https://sites.google.com/a/fe.unpad.ac.id/ekki/
– Submitting assignments online
» You need a gmail account and email me
Main Text
• Wooldridge, J.,
Introductory
Econometrics: A
Modern Approach,
2013
Today also includes:
Supplementary Text
• Zax, Jeffres S.,
Introductory
Econometrics:
Intuition, proof, and
practice, 2011
Supplementary Text
• Dougherty, C.,
Introduction to
Econometrics, 2011
• Gujarati, D., Basic
Econometrics, 2009
 Prerequisites
• Normally students taking econometrics
subjects must have taken
– Mathematics
– Statistics
– Microeconomics
– Macroeconomics

http://www.utexas.edu/cola/
depts/economics/
 Grading

• Tutorial  20 %
• Mid term exam  30%
• Final Exam  40%
• Assignments  10%
 Computer software issues
• We will use ??? in tutorials
– I recommend GRETL
– EVIEWS is good, but it is NOT a freeware
• However, EVIEWS, STATA (version 11 or 12)
and MS EXCEL may be used (especially in
lectures)
Econometrics

OVERVIEW
 Why econometrics prevails?
• A need in empirical evidence of economic the
ory
• Increasing use of mathematical methods in ec
onomics
• Advancement in computers and computing
• The lowering price (increasing supply) of econ
omic-related data
Theoretical vs empirical economics
• We know (from theories) that in certain
situation, subsidy is not only important but
also necessary to be distributed to the poor
– How effective is the subsidy?
• We know (from theories) that (basic)
infrastructure development is important for
economic growth
– How is the effect or impact of infrastructure
development on economic inequality?
13
Computing and computers
Free Economic Data
 “The ability to take data - to be able to understand it, to
process it, to extract value from it, to visualize it, to
communicate it, is going to be a hugely important skill in the
next decades…. Because now we really do have essentially free
and ubiquitous data. So the complimentary scarce factor is the
ability to understand that data and extract value from it.”
 What is Econometrics?
• The application of mathematical statistics to
economic data, to lend empirical support to
the models constructed by mathematical
economics, and to obtain numerical results

(G.S Madalla)

19
 Why are you studying econometrics?
• In order to be able to test economic theories
(empirical verification)
• In order to enhance economic theories based
on empirical study

20
 Causality and correlation
• Causality is probably the main reason why we
need econometrics
• Issues in establishing causality:
– Requires economic theory & econometric
assumptions to be satisfied
– Feedback/reverse causality
• Ceteris paribus assumption
 Causality and correlation
• “Correlation does not imply causality”
– 2 variables may be closely related (high correlation
coefficient)
– Does it mean that the change in one of the
variables causes the change in the other?
 Quantitative Analysis in Economics
• Econometrics is one of many tools in
conducting applied quantitative economic
research
• Other tools may be:
– Applied statistics
– Computable General Equilibrium Modelling
– etc
 The structure of economic data
• Cross Sectional Data
– consists of a sample of individuals, households,
firms, cities, states, countries, or a variety of other
units, taken at a given point in time
• Time Series Data
– Consists of observations on a variable or several
variables over time

24
 The structure of economic data
• Pooled Cross Section Data
– combination of two or more cross-sectional data

• Panel or Longitudinal Data

– A panel data set consists of a time series for each
cross-sectional member in the data set

25
 Obtaining Data
• Primary Data Collection
– Used mostly in applied microeconometrics (cross-
section type data)
– Uses of surveys, questionnaires, interviews
– Example: to analyse the effectiveness of
econometrics lecture at Unpad, we need to collect
data of the students

31
 Obtaining Data
• Secondary Data Collection
– Used in either applied microeconometrics or
macroeconometrics (mostly time-series type data)
– Can be bought from the sources
– Example: to analyse the effect of exchange rate
fluctuation on a country’s export performance, we
just need to get the data from the country’s
statistical bureau or from other organisations.

32
 Macro vs Micro Data
MACRO DATA MICRO DATA
• General Information • Specific information
– Aggregate – Individual level
– Household level
• Example:
– Firm level
– GDP of a country
– GDRP of a region
• Example:
– Individual characteristics such as
– Inflation in a country
gender, age, education
– Trade of countries – Firm characteristics such as
• Usually available in hard owner, labor, cost etc
copy • Available in raw data
• Number of observation is • Number of observation is large
small
 Indonesian Macro Data
• Statistik Ekonomi Keuangan Indonesia
• Worldbank data on Indonesia
• etc
 Indonesian Micro Data
• SUSENAS (Survey Sosial Ekonomi Nasional)
• SAKERNAS (Survey Angkatan Kerja Nasional)
• SURVEY INDUSTRI MANUFAKTUR
• INDONESIAN FAMILY LIFE SURVEY (IFLS)
• INDONESIAN DEMOGRAPHIC AND HEALTH
SURVEY (IDHS)
• Survey Pelayanan Kesehatan dan Pendidikan
• Survey Kualitas Pendidikan Anak
 Methodology of Econometrics
1. Statement of theory or hypothesis
2. Specification of the mathematical model of the
theory
3. Specification of the econometric model
4. Obtaining the data
5. Estimation of the parameters of the econometric
model
6. Hypothesis testing
7. Forecasting or prediction
8. Using the model for policy purposes

36
Model

FROM MATHEMATICAL TO
ECONOMETRICAL
Review: Functions Based on the number of
independent variables
• Univariate (one independent variable)
– Can be drawn in a 2D space graphs
– Qdx = f(Px) cet par
• Bivariate (two independent variables)
– Can be drawn in a 3D space graphs
– Q = f(K,L) cet par
• Multivariate (>2 independent variables)
– No Graphs

38
Multivariate General and specific function:
Another Example
• Variables that influence the demand for a laser
printer (QL) may be:
– The price of the printer (PL)  but not ONLY this one…
– The price of a computer (complementary product) (PC)
– The price of the toner (complementary product) (PT)
– The price of an ink jet printer (substitute product) (PI)
– Average income (Y)
– Advertising cost (A)
– Expected price of the printer in the near future (PE)
– Etc (such as: tastes, preferences)

39
 The General Demand Function
• Demand for a laser printer can be modeled
mathematically as a general demand function
QL=QL(PL, PC, PT, PI, Y, A, PE) cet.par.
• Using multivariable calculus (partial derivative)
we can analyze more about the change of an
independent variable and its effect on the
dependent variable

40
 The General Demand Function
QL=QL(PL, PC, PT, PI, Y, A, PE) cet.par
• How would you measure the change in QL
– If PL changes one unit?
– If Y changes one unit?
– If A changes one unit?

41
 The General Demand Function
• Using a stronger Ceteris Paribus assumption,
you may incorporate only PL, PC, and Y
QL=QL(PL, PC, Y) cet.par
• How would you measure the change in QL
– If PL changes one unit?
– If PC changes one unit?
– If Y changes one unit?

42
 The General Demand Function
• Using an even stronger Ceteris Paribus
assumption, you may incorporate only PL!
QL=QL(PL) cet.par
• You end up with a univariate demand function
• You still remember “the law of demand”
• You can draw a “demand curve”
– Using the inverse demand function

43
 Methodology of Econometrics
1. Statement of theory or hypothesis
2. Specification of the mathematical model of the
theory
3. Specification of the econometric model
4. Obtaining the data
5. Estimation of the parameters of the econometric
model
6. Hypothesis testing
7. Forecasting or prediction
8. Using the model for policy purposes

44
 Statement of theory or hypothesis
• According to microeconomic theory, variables that
influence the demand for a laser printer (QL) may be:
– The price of the printer (PL)  but not ONLY this one…
– The price of a computer (complementary product) (PC)
– The price of the toner (complementary product) (PT)
– The price of an ink jet printer (substitute product) (PI)
– Consumer’s income (Y)
– Advertising cost (A)
– Expected price of the printer in the near future (PE)
– Etc (such as: tastes, preferences)

45
Specification of the mathematical model of
the theory
• Demand for a laser printer can be modeled
mathematically as a general demand function
QL=QL(PL, PC, PT, PI, Y, A, PE) cet.par.
• Using multivariate calculus (partial derivative)
we can analyze more about the change of an
independent variable and its effect on the
dependent variable

46
 Specification of the
econometric model
• Assume that there is an exact (deterministic)
relationship between demand and the chosen
independent variables
• Assume the functional form of the demand function is
linear
• The deterministic linear demand function is

𝑳𝑷 𝑫𝒆𝒎𝒂𝒏𝒅𝒊=𝜷𝒐+𝜷𝟏 𝑳𝑷 𝑷𝑹𝑰𝑪𝑬𝒊+𝜷𝟐 𝑨𝑫𝑽 𝒊+𝜷𝟑 𝑪𝑶𝑴¿𝑷𝑹𝑰𝑪𝑬 𝒊+𝜷𝟒 𝑯𝑯𝑰𝑵𝑪𝒊+𝒖

 Specification of the
econometric model

𝑳𝑷 𝑫𝒆𝒎𝒂𝒏𝒅𝒊=𝜷𝒐+𝜷𝟏 𝑳𝑷 𝑷𝑹𝑰𝑪𝑬𝒊+𝜷𝟐 𝑨𝑫𝑽 𝒊+𝜷𝟑 𝑪𝑶𝑴¿𝑷𝑹𝑰𝑪𝑬 𝒊+𝜷𝟒 𝑯𝑯𝑰𝑵𝑪𝒊+𝒖

• All b s are the equation’s parameters:
– b0 is the constant term or intercept
– bi (i≠0) s are the coefficient of the variables
– u is “unobserved variables” or “error term”, or
“disturbance”, that is assumed to be random 
representing all other independent variables that are not
taken into account explicitly

48
 Specification of the
econometric model

𝑳𝑷 𝑫𝒆𝒎𝒂𝒏𝒅𝒊=𝜷𝒐+𝜷𝟏 𝑳𝑷 𝑷𝑹𝑰𝑪𝑬𝒊+𝜷𝟐 𝑨𝑫𝑽 𝒊+𝜷𝟑 𝑪𝑶𝑴¿𝑷𝑹𝑰𝑪𝑬 𝒊+𝜷𝟒 𝑯𝑯𝑰𝑵𝑪𝒊+𝝁

• Note that All b s and all independent variables are linear
• As can be seen, we just have an incomplete specific demand
function, because we don’t know the numerical values of the
parameters yet
• To get a specific demand function, we need to do 2 steps:
– Obtaining data (sample)
– Calculating the parameters
 The goal of econometric analysis
• The goal of any econometric analysis is to
estimate the parameters in the model and to
test hypotheses about this parameter

50
 Example:
Getting the Specific Demand Function
• Suppose that we have obtained a data set
containing the information regarding:
– Total sales of laser printer (reflecting quantity
demanded), in unit
– Price of laser printer, in million of rupiah
– Advertising cost in million of rupiah
– Price of computer in million of rupiah
– Income in million of rupiah

51
The Data

52
Getting the specific demand function
Getting the specific demand function
 The Specific Demand Function

• Hence, our specific demand function is:

^
𝑳𝑷 𝑫𝒆𝒎𝒂𝒏𝒅𝒊=𝟑𝟔𝟓.𝟓𝟒−𝟖𝟔.𝟖𝟖𝑳𝑷 𝑷𝑹𝑰𝑪𝑬𝒊+𝟔𝟒.𝟏𝟎𝑨𝑫𝑽 𝒊 −𝟏.𝟗𝑪𝑶𝑴 𝑷𝑹𝑰𝑪𝑬𝒊+𝟏𝟗.𝟗𝟐𝑯𝑯𝑰𝑵𝑪𝒊

55
 Example : modeling
• We want to know the
relationship between Subjective
Well Being and Income
• Even though what we are doing
is probably a ‘stylised fact’, we
need to have an economic
theory (or mixed with
psychology theory?) to explain
the relationship
– Also to find other variables
affecting SWB
 Model and Variables

We will use Linear Probability Model and Probit model to

estimate the equation
 The goal of econometric analysis
• The goal of any econometric analysis is to
estimate the parameters in the (economic)
model and to test hypotheses about this
parameter

58
Estimated econometric model
Econometric estimation - estimator
• Biasedness
• Efficiency
• Consistency