CHAPTER 1 Edited 2016
CHAPTER 1 Edited 2016
CHAPTER 1 Edited 2016
ECONOMETRICS
(Econ 3061)
1
Course Contents for Lecture:
Unit 1: Introduction ............................. 4 Hrs.
2
CHAPTER ONE
INTRODUCTION
1.1. What is Econometrics?
1.2. Goals/Uses of Econometrics
1.3 Models, Economic Models & Econometric
Models
1.4. The Econometric Approach
1.5. Division of Econometrics
1.6. Types of Data for Econometric Analysis
1.7. Causality & Ceteris Paribus
3
1.1 What Is Econometrics?
WHAT IS ECONOMETRICS?
It is economic measurement.
It is application of statistical methods to
economics.
It is application of mathematical &
statistical techniques to data in order to
collect evidence on questions of interest to
economics.
It combines economic theory,
mathematical economics, economic
4
statistics & mathematical statistics.
1.1 What Is Econometrics?
Economic Mathematical
theory economics
provides theory or expresses economic
imposes a logical theory using
structure on the mathematical form
question
Econometrics
data presentation
and description estimation and
testing techniques
Economic Mathematical
statistics statistics 5
1.2. GOALS/USES OF ECONOMETRICS
1. Estimation of economic parameters or
relationships needed for policy- or
decision-making;
2. Testing (and perhaps refining) economic
theory;
3. Forecasting/prediction of future values
of economic magnitudes; and
4. Evaluation of policies/programs.
6
1.3. Models, Economic Models & Econometric Models
Simplified representations of the real
world phenomena.
Models on relationships between/among
economic variables.
Economic models combined with assump-
tions about the random nature of the data
ECONOMETRIC
MODELS
ECONOMIC MODELS
MODELS 7
1.4. The Econometric Approach
1. Economic Theory/Model
2. Mathematical Model
3. Econometric Model (economic theory in an
empirically testable form)
+ Some a priori
information
4. Data
5. Estimation of the Model
6. Tests of Hypothesis (verifying any claim
suggested by the economic model)
7. Interpreting Results and Using the
Model for the Particular Purpose 8
1.4. The Econometric Approach
1. Statement of theory or hypothesis:
e.g.: People increase consumption as
income increases, but not by as much
as the increase in their income.
2. Specification of mathematical model:
C = α + βY; 0 < β < 1.
where: C = Consumption,
Y = Income,
β = slope = ΔC/ΔY,
α = intercept 9
1.4. The Econometric Approach
3. Specification of the econometric model:
C = α + βY + ɛ; 0 < β(= MPC) < 1.
α = intercept = autonomous consumption
ɛ = error/stochastic/disturbance term.
It captures several factors:
omitted variables,
measurement error in the dependent
variable and/or wrong functional form.
randomness of human behavior …
10
1.4. The Econometric Approach
4. Obtain data….
5. Estimate parameters of the model:
How? Various (3 broad) methods!
For now, suppose Cˆ i 184.08 0.8Yi
6. Hypothesis testing:
Is 0.8 statistically significantly > 0? < 1?
7. Interpret the results & use the model for
policy or forecasting:
1 Br in Y an 80 cent in average C.
If Y = 0, then average C = 184.08.
Pick a value of control variable (Y) to get
11
a desired value of target variable (C).
1.5. Division Of Econometrics
14
cont. …
In applied Econometrics we use the
tools of theoretical econometrics to
study some special field(s) of
economics, such as the production
function, consumption function,
investment function, demand and
supply functions, etc.
15
1.6. Types of Data for Econometric Analysis
Time series data: a set of observations on
the values that a variable takes at different
times. e.g. money supply, unemployment
rate, … over years.
Cross-sectional data: data on one/more
variables collected at a point in time.
Pooled data: cross-sectional observations
collected over time, but the units don’t
have to be the same.
Longitudinal/panel data: a special type of
pooled data in which the same cross-
sectional unit (say, a family or a firm) is
16
surveyed over time.
1.7. Causality & Ceteris Paribus
The goal of most of economic tests is to
infer that a variable has a causal effect on
another.
Finding an association between variables
might be suggestive, but is rarely sufficient.
The notion of ceteris paribus plays an
important role in causal analysis.
Because of the non-experimental nature of
most data in the social sciences,
uncovering causal relationships is very
challenging.
But, when carefully applied, econometric
methods can simulate a ceteris paribus
experiment. 17