Introduction
Introduction
econometrics
A short history
of
econometrics
Charles Adams
Semester 1, 2022
A short history
of
econometrics
3 Methodology of econometrics
A short history
of
econometrics Literal meaning 1: Economic measurement
Methodology
of
Literal meaning 2: Finding a model to summarize reality
econometrics
Literal meaning 3: Quantitative analysis of actual economic phenomena
Involving the following:
Economic theory
Mathematics
Statistics
Data
Your computer (Excel/E-Views/Stata/SPSS/R/Python)
A short history
of
econometrics
Methodology
of
econometrics
A short history
of
econometrics The early years:
Methodology
of
Developed as a separate discipline during the 1930s and 1940s
econometrics Early practitioners: Lawrence Klein and Jan Tinbergen
Early practical application constrained by lack of computing power
Cowles Commission methodology developed during 1950s and 1960s
Starting point: Classical Linear Regression Model; focus on error term assumptions
What to do if the assumptions don’t hold
Endogeneity; measurement error
Focus on fixing the regression results; underlying approach was not questioned
A short history
The 1960s and 1970s: era of large The 1980s and 1990s: A rethink of
of
econometrics
macro-models. econometric theory and
Methodology e.g. Wharton model in US - 2000 methodology.
of
econometrics equations.
Problems with traditional
BER and Treasury models in RSA.
econometrics: spurious
Used for policy analysis and correlation/regression.
forecasting.
Solution: cointegration
Economic instability of the 1970s
discredited large econometric Lucas critique
models’ forecasting ability.
Solution; micro-foundations,
Similar criticism to economic
structural DSGE models.
methodology arose post GFC.
ECO242 Chapter 0: History and methodology 6/21
Methodology of econometrics
What is
econometrics
A short history
of
econometrics 1 Statement of theory or hypothesis.
Methodology
of
2 Specification of the mathematical model of the theory.
econometrics
3 Specification of the statistical, or 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 control or policy purposes.
A short history
of
econometrics
Methodology
of Statement of the theory or hypothesis
econometrics
When (X) increases, consumption (Y ) increases, but the latter increase is not
as much as the former.
That is, the marginal propensity to consume (M P C) is greater than 0 but
smaller than 1.
A short history
of Mathematical model
econometrics
A short history
of
econometrics
Methodology
of
econometrics
Mathematical model
A short history
Econometric model
of
econometrics Relationship between income (X) and consumption (Y ) likely to be inexact.
Methodology
of
Therefore, add an error term (u) to the mathematical model.
econometrics
Consumption = β1 + β2 Income + u
Consumption
\ = β̂1 + β̂2 Income
Residuals (û) = Consumption − Consumption
\
The error term is a random (stochastic) variable that represents factors that
affect consumption but are not taken into account explicitly.
Figure 1.2
A short history
of
econometrics
Methodology
of
econometrics
Obtain data (usually from a sample)
Time-series (if looking at income vs. consumption in 1981-2000)
Cross-section (if examining income vs. consumption of 1000 employees in
Bellville in 2015)
A short history
of
econometrics
Methodology
of
econometrics
A short history
of
econometrics Estimation of the econometric model
Methodology
of Ordinary Least Squares (OLS)
econometrics
Derives estimates of the parameters by minimizing the sum of squared
residuals.
Example Consumption = β1 + β2 Income
Consumption
\ = −0.125 + 0.63Income
What do the parameter values, −0.125 and 0.63 mean?
A short history
of
econometrics
Methodology
of
econometrics
A short history
of Hypothesis testing
econometrics
Methodology
Is the (positive) relationship between income and consumption statistically
of significant?
econometrics
Moreover, is the estimated slope parameter statistically different from 1?
Statistical inference: Drawing conclusions about a population, based on a
sample of population
Sample regression line and parameters 6= Population regression line and
parameters
How good is the model?
Over-fitting, under-fitting, functional form, etc.
A short history
of
econometrics
Methodology
of
Forecasting or prediction
econometrics
Predicting the future value of the dependent variable on the basis of the
known value of the independent variable.
Example Consumption
\ = −0.125 + 0.63Income
If GDP is R5 trillion in 2020, what will consumption expenditure be?
A short history
of Use of model for control or policy purposes
econometrics
Methodology
If a certain level of Y is targeted, by how much should X be changed?
of
econometrics Example: Government wants to raise consumption expenditure to R2 trillion in
order to increase employment levels.
A short history
of
econometrics
Methodology
of
econometrics
End!
A short history
of
econometrics
Methodology
of
econometrics