Scribd
Upload
31 views32 pages

Chapter 2

Uploaded by

rafat.jallad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
31 views32 pages

Chapter 2

Uploaded by

rafat.jallad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 32

Applied Econometrics 4th edition

Dimitrios Asteriou
and
Stephen G Hall
Applied Econometrics 4th edition

INTRODUCTION

1. What is Econometrics?
2. The Stages of Econometric Work
Applied Econometrics 4th edition

What is Econometrics?

Econometrics means measurement (metrics in


greek) in economics.
The importance of applied work in economics is
increasing constantly.
Theory suggests that X affects Y but is this true
or not?
This is the work of the applied econometrician.
Applied Econometrics 4th edition

What is Econometrics?

• Examples of problems that may be tackled


by an Econometrician
(a) Modelling long-term relationships among
prices and interest rates.
(b) Examining the effect of inflation in
unemployment rates
(c) Examining the effect of disposable income on
consumption
Applied Econometrics 4th edition

What is Econometrics?
• Examples of problems that may be tackled by
an Econometrician (continued)
(a) Determining the factors that affect GDP per
capita growth
(b) Testing the validity of the CAPM and APT
theories
(c) Forecasting the correlation among the returns
and the stock indices of two countries.
Applied Econometrics 4th edition

The Stages of Applied


Econometric Analysis

Economic Theory

Econometric Model Data

Model Estimation

Specification Testing and Diagnostics

Is the Model Adequate?

No Yes

Test any hypothesis

Use for Predictions and Policy


Making
Applied Econometrics 4th edition

The Structure of Economic Data

• There are three different types of economic


data

• TIME SERIES
• CROSS SECTIONAL
• PANEL DATA
Applied Econometrics 4th edition

The Structure of Economic Data

• Time Series Data


• Examples: GDP, Unemployment, Inflation,
Stock Prices, etc
• Vectors or Columns (like in a spreadsheet)
• Frequencies: Yearly, Bi-annually, Quarterly,
Monthly, Weekly, Daily, Hourly.
• Lots of different values for different time
periods for one country, state, city, market.
Applied Econometrics 4th edition

The Structure of Economic Data

• Cross-Sectional Data
• Examples: GDP, Unemployment, Inflation,
Stock Prices, etc.
• Vectors or Rows (like in a spreadsheet)
• Frequencies: Yearly, Bi-annually, Quarterly,
Monthly, Weekly, Daily, Hourly.
• Lots of different values for different countries,
states, cities, markets, but for one time period
only.
Applied Econometrics 4th edition

The Structure of Economic Data


• Panel Data
• A combination of time series and cross-sectional data.
• Examples: GDP, Unemployment, Inflation, Stock Prices,
etc.
• Matrices (columns and rows to make an n times m matrix)
• Frequencies: Yearly, Bi-annually, Quarterly, Monthly,
Weekly, Daily, Hourly.
• Lots of different values for different countries, states, cities,
markets, and for different time periods.
Applied Econometrics 4th edition

The Structure of Economic Data - Notation

• Time series: Yt, t=1990, 1991, …, 2012


• Cross-Sectional: Yi, i=1, 2, 3, …, 40
• Panel Data: Yit, i and t defined as above.

• It is common to denote each observation by


the letter t and the total number of
observations by T for time series data, and to
denote each observation by the letter i and the
total number of observations by N for cross-
Applied Econometrics 4th edition

The Structure of Economic


Data – Quantitative vs Qualitative

• The data may be quantitative and qualitative.

• Quantitative (e.g. GDP per capita, exchange


rates, stock prices, unemployment rates)

• Qualitative (e.g. Day of the week, gender,


level of education)
12
Applied Econometrics 4th edition

Basic Data Handling

• Looking at raw data


• Graphical Analysis
• Summary Statistics
• Components of a Time Series
• Indices and Base Dates
• Data Transformations
Applied Econometrics 4th edition

Basic Data Handling - Looking at raw data

• Before getting into the statistical and


econometric tools, a preliminary analysis is
extremely important
• “Get the feel” of your data
• Look at the numbers on a spreadsheet. Note
number of series/end and start dates, range of
values etc.
• Outliers, discontinuities, structural breaks etc.
Applied Econometrics 4th edition

Basic Data Handling – Graphical Analysis


• Graphs facilitate the inspection process
• See the “big picture”
– Histograms: give an indication of the distribution of
a variable
– Scatter plots: give combinations of values from two
series for the purpose of determining their
relationship (if any)
– Line Graphs: facilitate the comparisons of series
– Bar Charts: good for comparisons
– Pie Charts: good for percentages/portions
Applied Econometrics 4th edition

Basic Data Handling – Histograms


• Histograms: give an indication of the distribution of a
variable
• Command in Eviews (View\Histogram and Stats)
Applied Econometrics 4th edition

Basic Data Handling – Scatter Plots


• Scatter Plots: give
combinations of values from
two series for the purpose of
determining their
relationship (if any)
• Eviews command (open the
two series together in a
group and choose
View/Graph/Scatter)
Applied Econometrics 4th edition

Basic Data Handling–Line Graphs

Line Graphs: facilitate


the comparisons of series

Command in Eviews:
Plot X Y
Applied Econometrics 4th edition

Basic Data Handling–Bar Charts

Bar Charts:
facilitate the
comparisons of series

Command in Eviews:
View/Graph/Bar
Applied Econometrics 4th edition

Basic Data Handling–Pie Charts

Pie Charts: Good for


proportions

Command in Eviews:
View/Graph/Pie
Applied Econometrics 4th edition

Basic Data Handling – Summary Statistics


• Summary statistics provide a more precise idea of the distribution of a variable
(mean, variance, st. dev. etc)
• For comparisons open the variables in a group.
Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

An economic or financial time series consists of up to four


components
1. Trend (smooth, long-term/consistent upward or downward
movement)
2. Cycle (rise and fall over periods longer than a year)
3. Seasonal (within year pattern seen in frequency data)
4. Irregular (random component, episodic – unpredictable but
identifiable – and residual – unpredictable and unidentifiable)
Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

An economic or financial time series consists of up to four components


Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

Seasonal (within year pattern seen in quarterly, monthly or weekly data)


Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

Trend (smooth, long-term/consistent upward or downward movement)


Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

Cycle (rise and fall over periods longer than a year)


Applied Econometrics 4th edition

Basic Data Handling–Components of a Time Series

Irregular (random component, episodic – unpredictable but


identifiable – and residual – unpredictable and unidentifiable)
Applied Econometrics 4th edition

Basic Data Handling – Indices and Base Dates

• Splicing two indices and change the base date

218 is 100
258 is X?
X=100*(258/218) or X=258*(100/218) or X=258/2.81
Applied Econometrics 4th edition

Basic Data Handling – Data Transformations

• Nominal vs Real data (use appropriate price deflator)


Applied Econometrics 4th edition

Basic Data Handling – Data Transformations

• Natural Logs (ln) have three advantages:

1. Natural logs linearize the exponential trends in the


series (makes the graph look smoother).
2. Natural logs linearize a model that is non linear in
parameters (consider Cobb-Douglas Production
function).
3. The OLS coefficients are elasticities.
Applied Econometrics 4th edition

Basic Data Handling – Data Transformations

• Differencing
In case we want to remove the trend component from a
time series entirely (i.e. in order to make it stationary)

First differences:
ΔYt = Yt – Yt-1
Second diffferences:
Δ2Yt = Δ(Yt – Yt-1) = ΔYt – ΔYt-1 =
= (Yt – Yt-1) – (Yt-1 – Yt-2) = Yt – 2Yt-1+ Yt-2
Applied Econometrics 4th edition

Basic Data Handling – Data Transformations

• Growth rates

Discretely compounded
growth rate of Yt =(Yt-Yt-1)/Yt-1

Continuously compounded
growth rate of Yt = ln(Yt-Yt-1) = ln(Yt)-ln(Yt-1)

You might also like